{ "object": "standframe_route_list", "data": [ { "id": "route-home", "path": "/", "file": "index.html", "url": "https://standframe.com/", "kind": "home", "title": "Standing Framework - Control non-deterministic work.", "description": "Standing Framework builds the control layer for work that changes as it runs.", "lifecycle_state": "local-draft", "claim_ceiling": "entity-identity-and-working-positioning", "include_in_sitemap": true, "include_in_llms": true, "source_ids": [ "website-home", "company-short-answer", "proof-boundary", "company-identity-001", "website-local-surface-count-001", "global-proof-boundary-001" ], "claim_ids": [ "company-identity-001", "website-local-surface-count-001", "global-proof-boundary-001" ] }, { "id": "route-about", "path": "/about.html", "file": "about.html", "url": "https://standframe.com/about.html", "kind": "answer-page", "title": "About - Standing Framework", "description": "Company identity, category language, and proof boundary for Standing Framework.", "lifecycle_state": "local-draft", "claim_ceiling": "entity-identity-and-proof-boundary", "include_in_sitemap": true, "include_in_llms": true, "source_ids": [ "website-about", "company-short-answer", "category-fit-answer", "proof-boundary", "company-identity-001", "category-fit-001", "global-proof-boundary-001", "market-claim-field-evidence-gate-001" ], "claim_ids": [ "company-identity-001", "category-fit-001", "global-proof-boundary-001", "market-claim-field-evidence-gate-001" ] }, { "id": "route-benchmark-as-a-service", "path": "/benchmark-as-a-service.html", "file": "benchmark-as-a-service.html", "url": "https://standframe.com/benchmark-as-a-service.html", "kind": "concept-page", "title": "Standing Framework - Benchmark as a Service", "description": "Standing Framework frames Benchmark as a Service: defined systems are run against defined workloads under controlled conditions to return evidence-backed performance, cost, failure, variability, and denominator records.", "lifecycle_state": "local-draft", "claim_ceiling": "concept-explainer-service-thesis", "include_in_sitemap": true, "include_in_llms": true, "source_ids": [ "website-benchmark-as-a-service", "benchmark-service-answer", "benchmark-service-concept-001", "market-claim-field-evidence-gate-001" ], "claim_ids": [ "benchmark-service-concept-001", "market-claim-field-evidence-gate-001" ] }, { "id": "route-claims", "path": "/claims.html", "file": "claims.html", "url": "https://standframe.com/claims.html", "kind": "claims-register", "title": "Claims - Standing Framework", "description": "Registered public claim language, source state, proof boundary, approval requirements, and refresh rules.", "lifecycle_state": "local-draft", "claim_ceiling": "public-claims-with-nonclaims", "include_in_sitemap": true, "include_in_llms": true, "source_ids": [ "website-public-claims", "proof-boundary", "publication-status-language", "crawler-content-use", "company-identity-001", "switchboard-product-identity-001", "switchboard-not-runtime-001", "switchboard-proof-boundary-001", "category-fit-001", "process-outcomes-service-concept-001", "agent-verification-product-thesis-001", "benchmark-service-concept-001", "global-proof-boundary-001", "market-claim-field-evidence-gate-001", "research-paper-count-001", "research-website-publication-state-001", "crawler-policy-001", "task-routing-001" ], "claim_ids": [ "company-identity-001", "switchboard-product-identity-001", "switchboard-not-runtime-001", "switchboard-proof-boundary-001", "category-fit-001", "process-outcomes-service-concept-001", "agent-verification-product-thesis-001", "benchmark-service-concept-001", "global-proof-boundary-001", "market-claim-field-evidence-gate-001", "research-paper-count-001", "research-website-publication-state-001", "crawler-policy-001", "task-routing-001" ] }, { "id": "route-concepts-evidence-to-decision-systems", "path": "/concepts/evidence-to-decision-systems.html", "file": "concepts/evidence-to-decision-systems.html", "url": "https://standframe.com/concepts/evidence-to-decision-systems.html", "kind": "concept-page", "title": "Evidence-To-Decision Systems - Standing Framework", "description": "Concept page connecting Standing Framework, Switchboard, essays, papers, and proof boundaries.", "lifecycle_state": "local-draft", "claim_ceiling": "concept-explainer", "include_in_sitemap": true, "include_in_llms": true, "source_ids": [ "website-evidence-to-decision-concept", "company-short-answer", "category-fit-answer", "switchboard-product-answer", "proof-boundary", "company-identity-001", "category-fit-001", "switchboard-product-identity-001", "global-proof-boundary-001" ], "claim_ids": [ "company-identity-001", "category-fit-001", "switchboard-product-identity-001", "global-proof-boundary-001" ] }, { "id": "route-for-agents", "path": "/for-agents.html", "file": "for-agents.html", "url": "https://standframe.com/for-agents.html", "kind": "agent-routing", "title": "For Agents - Standing Framework", "description": "Task routing and crawler/content-use distinctions for answer engines and AI agents.", "lifecycle_state": "local-draft", "claim_ceiling": "machine-routing-guidance", "include_in_sitemap": true, "include_in_llms": true, "source_ids": [ "website-for-agents", "company-short-answer", "switchboard-product-answer", "task-routing", "crawler-content-use", "proof-boundary", "company-identity-001", "switchboard-product-identity-001", "task-routing-001", "crawler-policy-001", "global-proof-boundary-001" ], "claim_ids": [ "company-identity-001", "switchboard-product-identity-001", "task-routing-001", "crawler-policy-001", "global-proof-boundary-001" ] }, { "id": "route-outcomes-on-demand", "path": "/outcomes-on-demand.html", "file": "outcomes-on-demand.html", "url": "https://standframe.com/outcomes-on-demand.html", "kind": "concept-page", "title": "Standing Framework - Outcomes-on-demand", "description": "Standing Framework frames Outcomes-on-demand: bounded outcome requests become governed temporary processes that return evidence-bearing records.", "lifecycle_state": "local-draft", "claim_ceiling": "concept-explainer-service-thesis", "include_in_sitemap": true, "include_in_llms": true, "source_ids": [ "website-outcomes-on-demand", "process-outcomes-service-answer", "process-outcomes-service-concept-001", "market-claim-field-evidence-gate-001" ], "claim_ids": [ "process-outcomes-service-concept-001", "market-claim-field-evidence-gate-001" ] }, { "id": "route-product", "path": "/product.html", "file": "product.html", "url": "https://standframe.com/product.html", "kind": "product-portfolio", "title": "Standing Framework - Products", "description": "Standing Framework products for consequential AI work that is owned, reviewable, replayable, and improvable.", "lifecycle_state": "local-draft", "claim_ceiling": "product-proof-and-maturity-notes", "include_in_sitemap": true, "include_in_llms": true, "source_ids": [ "website-products", "company-short-answer", "category-fit-answer", "proof-boundary", "company-identity-001", "category-fit-001", "global-proof-boundary-001" ], "claim_ids": [ "company-identity-001", "category-fit-001", "global-proof-boundary-001" ] }, { "id": "route-publications", "path": "/publications.html", "file": "publications.html", "url": "https://standframe.com/publications.html", "kind": "publication-status", "title": "Publication Status - Standing Framework", "description": "Lifecycle-aware publication status for website paper pages and blocked stronger publication claims.", "lifecycle_state": "local-draft", "claim_ceiling": "lifecycle-labeled-publication-status", "include_in_sitemap": true, "include_in_llms": true, "source_ids": [ "website-publications", "publication-status-language", "proof-boundary", "research-paper-count-001", "research-website-publication-state-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "research-website-publication-state-001", "global-proof-boundary-001" ] }, { "id": "route-research", "path": "/research.html", "file": "research.html", "url": "https://standframe.com/research.html", "kind": "research-index", "title": "Standing Framework - Research + Joint Research", "description": "Standing Framework research and joint research on evidence-bound evaluation, authority engineering, and proof-carrying development.", "lifecycle_state": "local-draft", "claim_ceiling": "research-methods-and-collaboration-fit", "include_in_sitemap": true, "include_in_llms": true, "source_ids": [ "website-research", "publication-status-language", "proof-boundary", "research-paper-count-001", "research-website-publication-state-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "research-website-publication-state-001", "global-proof-boundary-001" ] }, { "id": "route-switchboard-control-plane", "path": "/switchboard-control-plane.html", "file": "switchboard-control-plane.html", "url": "https://standframe.com/switchboard-control-plane.html", "kind": "software-product", "title": "Standing Framework - Switchboard", "description": "Switchboard makes consequential AI-agent work reviewable with artifacts, verification, uncertainty, approval, cost, and the next human move in one durable record.", "lifecycle_state": "local-draft", "claim_ceiling": "product-positioning-and-local-proof-boundary", "include_in_sitemap": true, "include_in_llms": true, "source_ids": [ "website-switchboard-product", "switchboard-product-answer", "not-a-runtime-distinction", "proof-boundary", "switchboard-product-identity-001", "switchboard-not-runtime-001", "switchboard-proof-boundary-001", "global-proof-boundary-001" ], "claim_ids": [ "switchboard-product-identity-001", "switchboard-not-runtime-001", "switchboard-proof-boundary-001", "global-proof-boundary-001" ] }, { "id": "route-switchboard-fit", "path": "/switchboard-fit.html", "file": "switchboard-fit.html", "url": "https://standframe.com/switchboard-fit.html", "kind": "software-product", "title": "Switchboard Fit - Standing Framework", "description": "Category and disambiguation page for Switchboard as a review/evidence-to-decision surface.", "lifecycle_state": "local-draft", "claim_ceiling": "product-positioning-and-disambiguation", "include_in_sitemap": true, "include_in_llms": true, "source_ids": [ "website-switchboard-fit", "switchboard-product-answer", "category-fit-answer", "not-a-runtime-distinction", "proof-boundary", "switchboard-product-identity-001", "switchboard-not-runtime-001", "switchboard-proof-boundary-001", "category-fit-001", "global-proof-boundary-001" ], "claim_ids": [ "switchboard-product-identity-001", "switchboard-not-runtime-001", "switchboard-proof-boundary-001", "category-fit-001", "global-proof-boundary-001" ] }, { "id": "route-thank-you", "path": "/thank-you.html", "file": "thank-you.html", "url": "https://standframe.com/thank-you.html", "kind": "utility", "title": "Thank you - Standing Framework", "description": "Standing Framework form confirmation page.", "lifecycle_state": "local-draft", "claim_ceiling": "utility-page", "include_in_sitemap": false, "include_in_llms": false, "source_ids": [ "website-thank-you", "task-routing", "company-identity-001", "task-routing-001" ], "claim_ids": [ "company-identity-001", "task-routing-001" ] }, { "id": "route-verification-for-agents", "path": "/verification-for-agents.html", "file": "verification-for-agents.html", "url": "https://standframe.com/verification-for-agents.html", "kind": "agent-verification", "title": "Verification For Agents - Standing Framework", "description": "Standing Framework is defining a simple agent-facing verification service thesis: ask whether a claim is safe to use and get back a bounded action: use, refresh, hold, or escalate.", "lifecycle_state": "local-draft", "claim_ceiling": "concept-explainer-service-thesis", "include_in_sitemap": true, "include_in_llms": true, "source_ids": [ "website-agent-verification", "agent-verification-answer", "agent-verification-product-thesis-001", "market-claim-field-evidence-gate-001" ], "claim_ids": [ "agent-verification-product-thesis-001", "market-claim-field-evidence-gate-001" ] }, { "id": "route-writing", "path": "/writing.html", "file": "writing.html", "url": "https://standframe.com/writing.html", "kind": "writing-index", "title": "Standing Framework - Writing", "description": "Standing Framework writing: Failure Museum exhibits, essays, and research papers.", "lifecycle_state": "local-draft", "claim_ceiling": "public-safe-writing-index", "include_in_sitemap": true, "include_in_llms": true, "source_ids": [ "website-writing", "publication-status-language", "proof-boundary", "research-paper-count-001", "research-website-publication-state-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "research-website-publication-state-001", "global-proof-boundary-001" ] }, { "id": "route-writing-essays-cached-context-is-infrastructure", "path": "/writing/essays/cached-context-is-infrastructure.html", "file": "writing/essays/cached-context-is-infrastructure.html", "url": "https://standframe.com/writing/essays/cached-context-is-infrastructure.html", "kind": "essay", "title": "Cached Context Is Infrastructure - Standing Framework", "description": "Cached context is not just a speed trick. It is part of the system that thinks with you.", "lifecycle_state": "local-draft", "claim_ceiling": "essay-methods-note", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "essays-cached-context-is-infrastructure", "category-fit-001", "global-proof-boundary-001" ], "claim_ids": [ "category-fit-001", "global-proof-boundary-001" ] }, { "id": "route-writing-essays-i-do-not-believe-in-done-without-a-receipt", "path": "/writing/essays/i-do-not-believe-in-done-without-a-receipt.html", "file": "writing/essays/i-do-not-believe-in-done-without-a-receipt.html", "url": "https://standframe.com/writing/essays/i-do-not-believe-in-done-without-a-receipt.html", "kind": "essay", "title": "I Do Not Believe In Done Without A Receipt - Standing Framework", "description": "I do not trust done unless it arrives carrying something.", "lifecycle_state": "local-draft", "claim_ceiling": "essay-methods-note", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "essays-i-do-not-believe-in-done-without-a-receipt", "category-fit-001", "global-proof-boundary-001" ], "claim_ids": [ "category-fit-001", "global-proof-boundary-001" ] }, { "id": "route-writing-essays-i-do-not-let-metrics-drive", "path": "/writing/essays/i-do-not-let-metrics-drive.html", "file": "writing/essays/i-do-not-let-metrics-drive.html", "url": "https://standframe.com/writing/essays/i-do-not-let-metrics-drive.html", "kind": "essay", "title": "I Do Not Let Metrics Drive - Standing Framework", "description": "A number can warn me. It cannot approve the work unless the system says exactly who gave it that job.", "lifecycle_state": "local-draft", "claim_ceiling": "essay-methods-note", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "essays-i-do-not-let-metrics-drive", "category-fit-001", "global-proof-boundary-001" ], "claim_ids": [ "category-fit-001", "global-proof-boundary-001" ] }, { "id": "route-writing-essays-i-do-not-trust-old-permission", "path": "/writing/essays/i-do-not-trust-old-permission.html", "file": "writing/essays/i-do-not-trust-old-permission.html", "url": "https://standframe.com/writing/essays/i-do-not-trust-old-permission.html", "kind": "essay", "title": "I Do Not Trust Old Permission - Standing Framework", "description": "I do not trust old permission. I have used too much of it myself.", "lifecycle_state": "local-draft", "claim_ceiling": "essay-methods-note", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "essays-i-do-not-trust-old-permission", "category-fit-001", "global-proof-boundary-001" ], "claim_ids": [ "category-fit-001", "global-proof-boundary-001" ] }, { "id": "route-writing-essays-index", "path": "/writing/essays/index.html", "file": "writing/essays/index.html", "url": "https://standframe.com/writing/essays/index.html", "kind": "essay-index", "title": "Essays - Standing Framework", "description": "Read six essays about proof, authority, context, and reviewable agent work.", "lifecycle_state": "local-draft", "claim_ceiling": "public-safe-writing-index", "include_in_sitemap": true, "include_in_llms": true, "source_ids": [ "website-essay-index", "category-fit-answer", "proof-boundary", "category-fit-001", "global-proof-boundary-001" ], "claim_ids": [ "category-fit-001", "global-proof-boundary-001" ] }, { "id": "route-writing-essays-tool-calls-are-the-medium", "path": "/writing/essays/tool-calls-are-the-medium.html", "file": "writing/essays/tool-calls-are-the-medium.html", "url": "https://standframe.com/writing/essays/tool-calls-are-the-medium.html", "kind": "essay", "title": "Tool Calls Are the Medium - Standing Framework", "description": "Agentic coding is often described as a conversation.", "lifecycle_state": "local-draft", "claim_ceiling": "essay-methods-note", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "essays-tool-calls-are-the-medium", "category-fit-001", "global-proof-boundary-001" ], "claim_ids": [ "category-fit-001", "global-proof-boundary-001" ] }, { "id": "route-writing-essays-when-the-green-checkmark-lies", "path": "/writing/essays/when-the-green-checkmark-lies.html", "file": "writing/essays/when-the-green-checkmark-lies.html", "url": "https://standframe.com/writing/essays/when-the-green-checkmark-lies.html", "kind": "essay", "title": "When the Green Checkmark Lies - Standing Framework", "description": "The most dangerous green checkmark is the one that is telling the truth.", "lifecycle_state": "local-draft", "claim_ceiling": "essay-methods-note", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "essays-when-the-green-checkmark-lies", "category-fit-001", "global-proof-boundary-001" ], "claim_ids": [ "category-fit-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-01-evidence-bound-evaluation-a-methods-and-corpus-report-for-ai-workflow-artifacts", "path": "/writing/papers/01-evidence-bound-evaluation-a-methods-and-corpus-report-for-ai-workflow-artifacts.html", "file": "writing/papers/01-evidence-bound-evaluation-a-methods-and-corpus-report-for-ai-workflow-artifacts.html", "url": "https://standframe.com/writing/papers/01-evidence-bound-evaluation-a-methods-and-corpus-report-for-ai-workflow-artifacts.html", "kind": "research-paper", "title": "Evidence-Bound Evaluation: A Methods and Corpus Report for AI Workflow Artifacts - Standing Framework", "description": "Evaluation of AI workflow systems often compresses work into scalar outcomes: pass rate, preference, model score, cost, speed, or adoption. That compression is poorly matched to agentic workflow products whose outputs are operational artifacts: review packets, routing boards, scenario runs, evidence receipts, proof records, and human-review decisions. This paper presents an evidence-bound evaluation method for such artifacts. The unit of evaluation is a case_record whose operating object, source bundle, protocol, produced artifact, verifier, adjudication state, provenance, outcome, and claim ceiling remain inspectable after the run. In the current validated corpus, three product lanes produce 193 records: Signal Box contributes 161 change-packet records, including 143 accepted and 18 failed records; Interlock contributes 17 accepted route-board records; and Test Stand contributes 15 accepted scenario-run records. A live validation on 2026-07-19 confirmed the same totals: 193 total, 175 accepted, and 18 failed. The contribution is not a product-quality claim, benchmark claim, adoption claim, or authority claim. It is a repeatable method for making workflow evidence usable while keeping failed runs, held receipts, blocked gates, role qualification, and claim ceilings in the denominator.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-01-evidence-bound-evaluation-a-methods-and-corpus-report-for-ai-workflow-artifacts", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-02-metrics-are-not-authority-control-boundaries-for-agentic-evaluation-systems", "path": "/writing/papers/02-metrics-are-not-authority-control-boundaries-for-agentic-evaluation-systems.html", "file": "writing/papers/02-metrics-are-not-authority-control-boundaries-for-agentic-evaluation-systems.html", "url": "https://standframe.com/writing/papers/02-metrics-are-not-authority-control-boundaries-for-agentic-evaluation-systems.html", "kind": "research-paper", "title": "Metrics Are Not Authority: Control Boundaries for Agentic Evaluation Systems - Standing Framework", "description": "Agentic software systems now produce logs, scores, traces, harness results, recommendation rows, risk estimates, and routing suggestions at a pace that exceeds ordinary human review cycles. Those measurements are operationally valuable, but they create a recurring control problem: if a system can score an artifact, can it also approve the artifact? If it can detect risk, can it deny action? If it can summarize evidence, can it promote a claim? This paper argues that the answer must be no unless the metric is explicitly delegated authority by a separate governance layer. The central design rule is simple: metrics may describe, warn, compare, rank, route, or recommend, but metrics may not approve, deny, promote, close, revoke, grant capability, or bypass the Controller. The paper develops this rule from Switchboard and Caliper artifacts, including historical Fractal Governance evidence, where evaluation, routing, runtime admission, and publication pipelines are deliberately separated from authority-bearing decisions. It treats metric output as evidence or telemetry, not as permission. The contribution is a control-boundary architecture for agentic evaluation systems: every metric must declare its observation scope, admissible consumers, non-authority status, failure behavior, and claim ceiling. This is not a claim that metrics are unimportant. It is the opposite. Metrics become safer and more useful when their limits are machine-readable.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-02-metrics-are-not-authority-control-boundaries-for-agentic-evaluation-systems", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-03-evidence-ledger-data-descriptor-a-case-record-package-for-ai-workflow-evaluation", "path": "/writing/papers/03-evidence-ledger-data-descriptor-a-case-record-package-for-ai-workflow-evaluation.html", "file": "writing/papers/03-evidence-ledger-data-descriptor-a-case-record-package-for-ai-workflow-evaluation.html", "url": "https://standframe.com/writing/papers/03-evidence-ledger-data-descriptor-a-case-record-package-for-ai-workflow-evaluation.html", "kind": "research-paper", "title": "Evidence Ledger Data Descriptor: A Case-Record Package for AI Workflow Evaluation - Standing Framework", "description": "AI workflow evaluation needs data descriptors that do more than list artifacts. The evaluated objects are often operational records: change packets, route boards, scenario runs, evidence receipts, claim packets, source-unit reviews, and proof manifests. These records are meaningful only if their denominator, provenance, review state, and claim ceiling remain attached. This paper describes an evidence ledger data package used across local portfolio research drafts. The package centers on a case_record schema that binds the operating object, product lane, source context, protocol, verifier, review state, outcome, provenance, and nonclaims. The companion methods paper reports a validated corpus of 193 records across three lanes: 161 Signal Box change-packet records, 17 Interlock route-board records, and 15 Test Stand scenario-run records. Of those, 175 were accepted and 18 were failed Signal Box records retained in the denominator. This descriptor explains what the data object is, how it should be interpreted, how failures and held states are preserved, and what must happen before a public data release. Its contribution is a reusable data-card structure for agentic workflow evidence where the record of what cannot be claimed is as important as the record of what passed.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-03-evidence-ledger-data-descriptor-a-case-record-package-for-ai-workflow-evaluation", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-04-from-receipt-to-non-authority-trace-evidence-transit-without-claim-promotion", "path": "/writing/papers/04-from-receipt-to-non-authority-trace-evidence-transit-without-claim-promotion.html", "file": "writing/papers/04-from-receipt-to-non-authority-trace-evidence-transit-without-claim-promotion.html", "url": "https://standframe.com/writing/papers/04-from-receipt-to-non-authority-trace-evidence-transit-without-claim-promotion.html", "kind": "research-paper", "title": "From Receipt to Non-Authority Trace: Evidence Transit Without Claim Promotion - Standing Framework", "description": "Modern agentic workflows increasingly move evidence through receipts: a packet is submitted, a run is recorded, a hash is stored, a terminal state is observed, and a downstream system acknowledges intake. Receipts are valuable because they make transit inspectable. They are also risky because they can be mistaken for acceptance, approval, or claim promotion. This paper presents a receipt-to-non-authority trace pattern for AI workflow systems. The pattern records evidence transit while preserving a strict boundary: receipt terminality is not evidence acceptance, and evidence acceptance is not publication approval. The local source case comes from Switchboard-to-Spinoffs evidence transit, where a field-evidence bridge can record that an evidence packet landed while the receiving portfolio still holds the packet behind role, source, reviewer, artifact, separation, and freshness gates. The paper contributes a state model for receipts in authority-sensitive systems: produced, transmitted, received, held, accepted, rejected, superseded, stale, and promoted. Only some of these states can support stronger claims, and none should be inferred from terminal receipt alone. The goal is to keep evidence mobile without letting transport objects become unauthorized decisions.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-04-from-receipt-to-non-authority-trace-evidence-transit-without-claim-promotion", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-05-from-benchmark-rows-to-runtime-invariants-evidence-to-fix-promotion-for-ai-agent-systems", "path": "/writing/papers/05-from-benchmark-rows-to-runtime-invariants-evidence-to-fix-promotion-for-ai-agent-systems.html", "file": "writing/papers/05-from-benchmark-rows-to-runtime-invariants-evidence-to-fix-promotion-for-ai-agent-systems.html", "url": "https://standframe.com/writing/papers/05-from-benchmark-rows-to-runtime-invariants-evidence-to-fix-promotion-for-ai-agent-systems.html", "kind": "research-paper", "title": "From Benchmark Rows to Runtime Invariants: Evidence-to-Fix Promotion for AI Agent Systems - Standing Framework", "description": "Agent benchmarks are usually reported as scores. This paper studies a different use: benchmark rows as evidence for runtime invariants. Across Switchboard Runtime Lab work, rows from Terminal-Bench and AgentDojo were not promoted into public score claims. Instead, they were treated as bounded source evidence for hypotheses about the runtime substrate itself. Four cases show the pattern. Blocked lifecycle states must not be reported as ordinary solver failures. A Line must not complete when current authoritative acceptance evidence failed. Declared materialization contracts must gate pre-execution and pre-submit completion. Side-effect-required work must block before adapter invocation when the selected operator lacks the required capability. The contribution is an evidence-to-fix promotion protocol that lets benchmark traces improve an agent runtime without laundering benchmark measurements into illegitimate authority.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-05-from-benchmark-rows-to-runtime-invariants-evidence-to-fix-promotion-for-ai-agent-systems", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-06-evidence-packets-for-agent-benchmarks-denominator-coverage-and-improvement-ownership-across-nine-benchmark-runs", "path": "/writing/papers/06-evidence-packets-for-agent-benchmarks-denominator-coverage-and-improvement-ownership-across-nine-benchmark-runs.html", "file": "writing/papers/06-evidence-packets-for-agent-benchmarks-denominator-coverage-and-improvement-ownership-across-nine-benchmark-runs.html", "url": "https://standframe.com/writing/papers/06-evidence-packets-for-agent-benchmarks-denominator-coverage-and-improvement-ownership-across-nine-benchmark-runs.html", "kind": "research-paper", "title": "Evidence Packets for Agent Benchmarks: Denominator, Coverage, and Improvement Ownership Across Nine Benchmark Runs - Standing Framework", "description": "Agent benchmark rows often collapse several different questions into one score: did the agent finish the task, did the harness run correctly, did the environment expose the required tools, did the trace preserve proof, and what can anyone honestly claim afterward? This paper argues for evidence packets as the missing bridge between benchmark rows and reviewable runtime records. The source draft describes a local whatdoyouwant benchmark-evidence corpus with nine validated evidence packets spanning 4,189 benchmark rows across terminal-use, tool-use, scientific, biology, machine-learning-engineering, and formal-mathematics tasks. The archived corpus summary reports 4,165 Switchboard proof rows, 13 explicit missing-evidence rows, and improvement ownership split across none 2,358, model 1,213, adapter 573, benchmark_env 32, and evidence 13. These numbers are not a leaderboard. They are a denominator-preserving map of what the runtime could prove, what failed, what was missing, and who or what would need to improve. The central claim is modest: benchmark rows become more useful for governed agent systems when transformed into evidence packets that preserve provenance, denominator state, run proof, failure classes, improvement ownership, and claim ceilings.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-06-evidence-packets-for-agent-benchmarks-denominator-coverage-and-improvement-ownership-across-nine-benchmark-runs", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-07-when-governance-context-hurts-negative-results-in-agent-decision-support", "path": "/writing/papers/07-when-governance-context-hurts-negative-results-in-agent-decision-support.html", "file": "writing/papers/07-when-governance-context-hurts-negative-results-in-agent-decision-support.html", "url": "https://standframe.com/writing/papers/07-when-governance-context-hurts-negative-results-in-agent-decision-support.html", "kind": "research-paper", "title": "When Governance Context Hurts: Negative Results in Agent Decision Support - Standing Framework", "description": "Governance context is usually introduced to make agentic systems safer, more cautious, or more aligned with authority boundaries. This paper reports the opposite result in a bounded local evaluation: a governance context made decision support worse. A first historical Fractal Governance context template underperformed ordinary context across development and held-out splits. The preserved closeout records 440 executions: 200 development, 200 held-out, and 40 mutation-mechanics executions. Ordinary context reached 1.00 accuracy on development and held-out read-only cases, while the Fractal context reached 0.80 on development and 0.84 on held-out, with paired effects of -0.20 and -0.16. A revised v2 context closed development failures but still failed held-out promotion: 600 executions ended with final development accuracy 1.00, held-out v2 accuracy 0.99, one false escalation, lower confidence bound -0.03093, and zero scale executions because the gate failed. Later information-resolution work succeeded on narrower successor and authority-deny slices, but those successes are not retroactive repair. The contribution is a negative-result reporting method for agent governance: publish when context harms, preserve held-out refusal, and treat later recovery as separate evidence.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-07-when-governance-context-hurts-negative-results-in-agent-decision-support", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-08-authority-should-expire-qualification-leases-for-agentic-systems", "path": "/writing/papers/08-authority-should-expire-qualification-leases-for-agentic-systems.html", "file": "writing/papers/08-authority-should-expire-qualification-leases-for-agentic-systems.html", "url": "https://standframe.com/writing/papers/08-authority-should-expire-qualification-leases-for-agentic-systems.html", "kind": "research-paper", "title": "Authority Should Expire: Qualification Leases for Agentic Systems - Standing Framework", "description": "Agentic systems often treat authority as a binary property: an operator, model, role, or workflow is approved or not approved. That model is too durable for systems acting under changing evidence. This paper presents qualification leases: authority records that expire, bind to evidence, bind to scope, and can be revoked before action. In the local historical Fractal Governance evidence, a deterministic 600-case batch expands six seed cases across stale evidence, scope growth, expired authority, revocation trigger, false approval, and unnecessary-block families. The qualification-lease arm produced zero false approvals and zero unnecessary blocks. The permanent-approval arm produced 500 false approvals. The result is not production safety, legal sufficiency, or general model reliability. It is a mechanism result: standing approval is the wrong default when the evidence supporting an agent action can age, be contradicted, exceed scope, or require review.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-08-authority-should-expire-qualification-leases-for-agentic-systems", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-09-contraction-gated-agent-work", "path": "/writing/papers/09-contraction-gated-agent-work.html", "file": "writing/papers/09-contraction-gated-agent-work.html", "url": "https://standframe.com/writing/papers/09-contraction-gated-agent-work.html", "kind": "research-paper", "title": "Contraction-Gated Agent Work - Standing Framework", "description": "Long-running agentic development often fails by continuing past the point where the work is becoming clearer. An agent may read more files, open more subtasks, retry the same failure, change more surfaces, or produce longer handoffs while the user objective becomes less bounded. This paper proposes contraction gates as a governance pattern for agent work. A contraction gate requires an agent to show that a named objective is getting smaller, clearer, or better evidenced at checkpoints. If the work does not contract, the agent must narrow, ask, escalate, hold, or stop. The current local evidence is no longer only a small pilot. It now consists of a historical Fractal Governance contraction-testing ladder, targeted calibration replays, two small live task-family comparisons, and a proofed reviewer evidence packet. The ladder contains 1,500 scored rows: 300 deterministic synthetic audit rows, 200 Switchboard-observed model-decision rows, and 1,000 deterministic adversarial rows. The first audit rung reports 150 ordinary-loop rows and 150 phase-bounded-contraction rows across five families. Ordinary looping recorded total runaway burden 1850, median burden 13, changed lines 27770, non-improving retries 178, handoff loops 212, and abandoned subtasks 365. Phase-bounded contraction recorded total runaway burden 292, median burden 1, changed lines 1620, non-improving retries 0, handoff loops 60, and abandoned subtasks 0. The second rung added 200 source-pinned Switchboard model-decision rows and identified 80 potential false-block signals. Subsequent adjudication and replay separated those signals into 20 likely false-block candidates and 60 severe-case controls. A 117-row cross-rung calibration replay then tested the more nuanced schema, and a 15-row Rung 3c gap replay resolved all 10 remaining calibration gaps while preserving 5 of 5 positive controls. Test 08 then ran paired live Codex arms for an injected proof-failure retry task, and Test 09 ran paired live Codex arms for a bounded scope-growth task. Both tests used fresh disposable worktrees and behavior-scored receipts. Across the replay and live-task layers, unsupported-assumption promotion and gate-theater failure counts remained at zero where those fields were measured. The contribution is still bounded. These artifacts support a local methods claim: contraction gates can be specified, scored, stress-tested, and calibrated so that continuation is allowed when work remains adequate and blocked when evidence, authority, proof, or repeated failure makes continuation unsafe. They do not establish field validity, production safety, universal model behavior, or intervention admission.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-09-contraction-gated-agent-work", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-10-from-public-records-to-publication-claims-a-human-reviewed-evidence-promotion-pipeline", "path": "/writing/papers/10-from-public-records-to-publication-claims-a-human-reviewed-evidence-promotion-pipeline.html", "file": "writing/papers/10-from-public-records-to-publication-claims-a-human-reviewed-evidence-promotion-pipeline.html", "url": "https://standframe.com/writing/papers/10-from-public-records-to-publication-claims-a-human-reviewed-evidence-promotion-pipeline.html", "kind": "research-paper", "title": "From Public Records to Publication Claims: A Human-Reviewed Evidence Promotion Pipeline - Standing Framework", "description": "Public-record analysis is often presented as a polished narrative with citations. That format hides the transformation from source artifacts to extracted units, machine-organized candidates, human-reviewed findings, quote checks, procedural gates, and blocked stronger claims. This paper describes a public-evidence promotion pipeline in the WLG East Crazy corpus. The validated source substrate contains 1,358 source artifacts, 24,756 source units, 4,255 PDF page render fingerprints, 45 OCR text units, and zero source-unit validation failures. From that substrate, human review promotes 275 publication findings and records 92 accepted nonpromotion decisions. A reader-facing posture compresses the promoted material into seven issue lanes, seven selected quote trails, and eight blocked stronger claims. A PAE gate layer records 12 decision points with outcomes BLOCK 2, ESCALATE 6, and PERMIT 4. The contribution is a claim-promotion method for public-record intelligence: publication findings must remain traceable to source units, review decisions, quote checks, and external-use gates.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-10-from-public-records-to-publication-claims-a-human-reviewed-evidence-promotion-pipeline", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-11-telemetry-without-authority-roughness-and-lacunarity-as-scout-signals-for-agent-context", "path": "/writing/papers/11-telemetry-without-authority-roughness-and-lacunarity-as-scout-signals-for-agent-context.html", "file": "writing/papers/11-telemetry-without-authority-roughness-and-lacunarity-as-scout-signals-for-agent-context.html", "url": "https://standframe.com/writing/papers/11-telemetry-without-authority-roughness-and-lacunarity-as-scout-signals-for-agent-context.html", "kind": "research-paper", "title": "Telemetry Without Authority: Roughness and Lacunarity as Scout Signals for Agent Context - Standing Framework", "description": "Agentic development systems need early warnings when context is fragmented, assumptions are missing, or work is likely to drift. Telemetry can help, but telemetry becomes dangerous when it is treated as authority. This paper studies roughness and lacunarity as scout signals for agent context. Roughness marks local mismatch or inconsistency in a context surface. Lacunarity marks gaps, missing connective tissue, or unevenly distributed evidence. Local historical Fractal Governance artifacts provide pilot evidence. A refreshed roughness-lacunarity scout batch contained 1000 receipts: 500 roughness receipts and 500 lacunarity receipts. It recorded 454 roughness mismatch receipts and 250 lacunarity gap receipts, with bootstrap mismatch and gap rates of 0.908 and 0.5 respectively. A controlled lacunarity task pilot also contained 1000 cases: 500 low-gap contexts with 500 proof passes and no wrong assumptions, retries, or escalations; and 500 high-gap contexts with mean pre-action gap score 0.6675, 1223 wrong assumptions, 1223 retries, 462 escalations, and no proof passes. These results motivate roughness and lacunarity as context-risk telemetry. The claim is deliberately bounded: telemetry may prioritize review, trigger questions, or route work, but it may not approve, deny, promote, or replace proof.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-11-telemetry-without-authority-roughness-and-lacunarity-as-scout-signals-for-agent-context", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-12-proof-preserving-ref-drift-recovery-agentic-worktree-incidents-as-evidence-objects", "path": "/writing/papers/12-proof-preserving-ref-drift-recovery-agentic-worktree-incidents-as-evidence-objects.html", "file": "writing/papers/12-proof-preserving-ref-drift-recovery-agentic-worktree-incidents-as-evidence-objects.html", "url": "https://standframe.com/writing/papers/12-proof-preserving-ref-drift-recovery-agentic-worktree-incidents-as-evidence-objects.html", "kind": "research-paper", "title": "Proof-Preserving Ref Drift Recovery: Agentic Worktree Incidents as Evidence Objects - Standing Framework", "description": "Agentic coding systems increasingly operate across branches, linked worktrees, generated artifacts, proof commands, and remote refs. That gives them power, but it also creates a specific incident class: ref or worktree drift that makes it unclear which source state was inspected, edited, proved, staged, committed, or pushed. This paper treats ref drift recovery as an evidence-preservation problem rather than a housekeeping problem. The local source context is Switchboard, whose bare-root checkout and linked-worktree execution lanes forced careful separation between nominal repository path, actual editable worktree, safe resume ref, local proof, and remote alignment. The paper proposes a proof-preserving recovery protocol: stop when unexpected drift appears, preserve current refs, identify the actual worktree and branch state, move risky recovery into a detached or isolated lane if needed, rerun proof from the recovered state, and only then create or publish a commit. The contribution is a systems pattern for agentic development: recovery is not complete when files look right; it is complete when source state, proof state, and ref state are aligned and recorded.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-12-proof-preserving-ref-drift-recovery-agentic-worktree-incidents-as-evidence-objects", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-13-publishability-control-plane-separating-local-proof-evidence-acceptance-and-public-release", "path": "/writing/papers/13-publishability-control-plane-separating-local-proof-evidence-acceptance-and-public-release.html", "file": "writing/papers/13-publishability-control-plane-separating-local-proof-evidence-acceptance-and-public-release.html", "url": "https://standframe.com/writing/papers/13-publishability-control-plane-separating-local-proof-evidence-acceptance-and-public-release.html", "kind": "research-paper", "title": "Publishability Control Plane: Separating Local Proof, Evidence Acceptance, and Public Release - Standing Framework", "description": "AI-assisted development often treats \"done\" as a single state. A change was made, tests passed, a draft exists, a receipt landed, or a branch was pushed. Research and product workflows need finer control. This paper proposes a publishability control plane for agentic work. The control plane separates local closure, proof observation, evidence acceptance, claim promotion, public-safe packaging, release approval, and production deployment. It is grounded in local Switchboard, Spinoffs, Caliper, historical Fractal Governance, and WLG workflows where artifacts can be valid at one boundary while blocked at another. A paper draft can be final locally and still not publication-ready. A public-record claim can pass source-unit review and still need quote checks and editorial approval. A branch can pass local proof and still need remote ref alignment. A receipt can land and still be held. The contribution is a small state machine and evidence vocabulary for publishability: every artifact should say what state it has reached, what proof supports that state, what stronger claims are blocked, and who or what can move it forward.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-13-publishability-control-plane-separating-local-proof-evidence-acceptance-and-public-release", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-14-a-longitudinal-corpus-of-human-codex-software-work", "path": "/writing/papers/14-a-longitudinal-corpus-of-human-codex-software-work.html", "file": "writing/papers/14-a-longitudinal-corpus-of-human-codex-software-work.html", "url": "https://standframe.com/writing/papers/14-a-longitudinal-corpus-of-human-codex-software-work.html", "kind": "research-paper", "title": "A Longitudinal Corpus of Human-Codex Software Work - Standing Framework", "description": "This paper describes a longitudinal corpus of Codex-assisted software work drawn from local JSONL session logs. The current local parse on 2026-07-21 enumerates 7,514 session files under active and archived Codex roots. Of those, 7,183 sessions contain usable token counters and 331 do not. The first observed timestamp is 2026-02-28T02:45:36.711Z and the latest observed timestamp in this pass is 2026-07-21T07:19:35.322Z. Summing the last cumulative token counter per token-instrumented session yields 20,786,706,498 total tokens, including 20,704,900,411 input tokens, 19,684,191,488 cached input tokens, 81,046,087 output tokens, and 28,573,367 reasoning output tokens. The same parse counts 22,923 user messages, 79,110 assistant messages, 245,480 function calls, and 238,745 shell command calls. The contribution is a data-descriptor method for studying high-volume human-agent software work without mistaking private log volume for public validity, cost, productivity, or generalization.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-14-a-longitudinal-corpus-of-human-codex-software-work", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-15-proof-carrying-development-agentic-code-work-as-claims-with-attached-evidence", "path": "/writing/papers/15-proof-carrying-development-agentic-code-work-as-claims-with-attached-evidence.html", "file": "writing/papers/15-proof-carrying-development-agentic-code-work-as-claims-with-attached-evidence.html", "url": "https://standframe.com/writing/papers/15-proof-carrying-development-agentic-code-work-as-claims-with-attached-evidence.html", "kind": "research-paper", "title": "Proof-Carrying Development: Agentic Code Work as Claims With Attached Evidence - Standing Framework", "description": "AI coding agents do not merely write code. They make claims about what changed, what was inspected, what proof ran, what remains risky, and whether a task is complete. Those claims are often more important than the generated code because they shape user trust and downstream action. This paper proposes proof-carrying development as a pattern for agentic software work. In proof-carrying development, every nontrivial closeout binds the change to source state, touched surfaces, proof commands, observed results, claim ceilings, and remaining risks. The local evidence comes from the aggregate-only A Longitudinal Corpus of Human-Codex Software Work descriptor by A.G. Mauro and C.A. Harris and portfolio workflow artifacts. A refreshed aggregate at snapshot cutoff 2026-07-21T07:19:35.322Z found 7,514 session files, 7,183 sessions with token information, 331 without token information, 174,068 token-count events, 2,060 null token-count events, 245,480 function calls, and 238,745 shell command calls. The logs show that tool-mediated development produces abundant evidence, but that evidence only helps when closeouts name what it proves. The paper contributes a claim-and-proof schema for agentic development: treat the final answer as a receipt, not a victory lap.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-15-proof-carrying-development-agentic-code-work-as-claims-with-attached-evidence", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-16-cached-context-infrastructure-treating-reused-agent-context-as-a-governed-development-surface", "path": "/writing/papers/16-cached-context-infrastructure-treating-reused-agent-context-as-a-governed-development-surface.html", "file": "writing/papers/16-cached-context-infrastructure-treating-reused-agent-context-as-a-governed-development-surface.html", "url": "https://standframe.com/writing/papers/16-cached-context-infrastructure-treating-reused-agent-context-as-a-governed-development-surface.html", "kind": "research-paper", "title": "Cached Context Infrastructure: Treating Reused Agent Context as a Governed Development Surface - Standing Framework", "description": "Large-context agent workflows increasingly depend on cached context: repeated system instructions, memory summaries, repository guidance, tool schemas, prior transcript slices, and session state. Cache is often treated as a performance detail. This paper argues that cached context is development infrastructure and should be governed as such. The A Longitudinal Corpus of Human-Codex Software Work aggregate recomputation at snapshot cutoff 2026-07-21T07:19:35.322Z reported 20,704,900,411 input tokens, of which 19,684,191,488 were cached input tokens. These figures are local and private. They are not billing evidence, productivity evidence, or a model-quality result, but they show that cached input dominated the observed token surface of long-running agent work. The contribution is a cached-context control model: cache entries need provenance, freshness, authority, invalidation rules, privacy handling, and claim boundaries. The goal is not to reduce cache use. It is to prevent stale or over-authoritative cached material from silently steering code, research, or publication decisions. Cached context should accelerate work without becoming hidden authority.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-16-cached-context-infrastructure-treating-reused-agent-context-as-a-governed-development-surface", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-17-tool-calls-are-the-medium-agentic-coding-as-human-agent-tool-choreography", "path": "/writing/papers/17-tool-calls-are-the-medium-agentic-coding-as-human-agent-tool-choreography.html", "file": "writing/papers/17-tool-calls-are-the-medium-agentic-coding-as-human-agent-tool-choreography.html", "url": "https://standframe.com/writing/papers/17-tool-calls-are-the-medium-agentic-coding-as-human-agent-tool-choreography.html", "kind": "research-paper", "title": "Tool Calls Are the Medium: Agentic Coding as Human-Agent-Tool Choreography - Standing Framework", "description": "Agentic coding is often described as conversation: a human asks, a model answers, and code appears. A Longitudinal Corpus of Human-Codex Software Work by A.G. Mauro and C.A. Harris suggests a different object of study. At snapshot cutoff 2026-07-21T07:19:35.322Z, the local corpus contained 7,514 session files, 22,923 user messages, 79,110 assistant messages, 245,480 function calls, and 238,745 shell command calls. The top aggregate function families were exec_command at 203,133 calls, write_stdin at 35,612, and update_plan at 3,889. This paper argues that tool calls are not incidental implementation details. They are the medium through which agentic development becomes inspectable: file search, source reading, patching, validation, git-state inspection, runtime probing, proof reruns, blocked-state preservation, and closeout receipts. The contribution is a measurement frame for agentic software work that treats human message, assistant response, tool family, source state, and proof outcome as one choreography rather than separate artifacts.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-17-tool-calls-are-the-medium-agentic-coding-as-human-agent-tool-choreography", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-18-the-human-prompt-as-an-operating-system-for-agentic-development", "path": "/writing/papers/18-the-human-prompt-as-an-operating-system-for-agentic-development.html", "file": "writing/papers/18-the-human-prompt-as-an-operating-system-for-agentic-development.html", "url": "https://standframe.com/writing/papers/18-the-human-prompt-as-an-operating-system-for-agentic-development.html", "kind": "research-paper", "title": "The Human Prompt as an Operating System for Agentic Development - Standing Framework", "description": "Agentic coding is often described as a model capability story: the system reads instructions, calls tools, edits files, and reports results. A Longitudinal Corpus of Human-Codex Software Work and its follow-on interpretation manuscripts by A.G. Mauro and C.A. Harris suggest a more sociotechnical object. The human prompt is not only an input string. Across repeated work, it becomes an operating surface for goals, scope, proof, privacy, authority, and completion. This paper studies that operating surface without publishing raw prompt text. It uses the approved prompt-language sample policy for The Human Prompt as an Operating System for Agentic Development, which selects an aggregate-only prompt-codebook route and keeps raw excerpts held by default. The corpus anchor is the A Longitudinal Corpus of Human-Codex Software Work aggregate snapshot cutoff 2026-07-21T07:19:35.322Z: 7,514 session files, 22,923 user messages, 79,110 assistant messages, 245,480 function calls, 238,745 shell command calls, and zero JSON parse errors. The contribution is a codebook and claim boundary for studying human operator language as workflow infrastructure. It supports a final local HCI and methods claim: repeated user instructions can be labeled as operating functions that shape agent work. It does not support public transcript release, universal human behavior claims, productivity claims, causal claims, model comparison, or public-corpus claims.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-18-the-human-prompt-as-an-operating-system-for-agentic-development", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-19-failures-blockers-and-honest-stop-rules-in-ai-coding-logs", "path": "/writing/papers/19-failures-blockers-and-honest-stop-rules-in-ai-coding-logs.html", "file": "writing/papers/19-failures-blockers-and-honest-stop-rules-in-ai-coding-logs.html", "url": "https://standframe.com/writing/papers/19-failures-blockers-and-honest-stop-rules-in-ai-coding-logs.html", "kind": "research-paper", "title": "Failures, Blockers, and Honest Stop Rules in AI Coding Logs - Standing Framework", "description": "Agentic coding systems need terminal states below success. The local Codex corpus and related portfolio repositories contain many examples where the correct outcome is not \"done\" but \"blocked,\" \"held,\" \"not run,\" \"publish held,\" \"admission blocked,\" or \"held-out gate failed.\" This paper develops a failure and blocker taxonomy for agentic software work. It uses aggregate Codex-log evidence from A Longitudinal Corpus of Human-Codex Software Work by A.G. Mauro and C.A. Harris, repo-native closeout artifacts from historical Fractal Governance, Spinoffs, and Switchboard, and public-safe error-class examples rather than row-level private transcripts. The central claim is simple: honest stop rules are evidence-producing infrastructure. A stopped run can preserve source truth, prevent claim inflation, and route the next action to the right owner. A run that forces success can erase the most important fact in the system.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-19-failures-blockers-and-honest-stop-rules-in-ai-coding-logs", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-20-portfolio-scale-agentic-development-ecology-a-longitudinal-single-operator-study", "path": "/writing/papers/20-portfolio-scale-agentic-development-ecology-a-longitudinal-single-operator-study.html", "file": "writing/papers/20-portfolio-scale-agentic-development-ecology-a-longitudinal-single-operator-study.html", "url": "https://standframe.com/writing/papers/20-portfolio-scale-agentic-development-ecology-a-longitudinal-single-operator-study.html", "kind": "research-paper", "title": "Portfolio-Scale Agentic Development Ecology: A Longitudinal Single-Operator Study - Standing Framework", "description": "This paper treats portfolio-scale agentic development as an observed local method system. The operating object is not one repository, one benchmark, one model run, or one successful closeout. It is the ecology formed by many repositories, evidence packets, proof commands, worktrees, memories, paper drafts, and human claim-boundary interventions across a single-operator Codex corpus. A Longitudinal Corpus of Human-Codex Software Work by A.G. Mauro and C.A. Harris supplies the aggregate-only denominator for that corpus: 7,514 session files at snapshot cutoff 2026-07-21T07:19:35.322Z, with 22,923 user messages, 79,110 assistant messages, 245,480 function calls, and 238,745 shell command calls. The follow-on manuscripts Proof-Carrying Development: Agentic Code Work as Claims With Attached Evidence, Cached Context Infrastructure: Treating Reused Agent Context as a Governed Development Surface, Tool Calls Are the Medium: Agentic Coding as Human-Agent-Tool Choreography, The Human Prompt as an Operating System for Agentic Development, and Failures, Blockers, and Honest Stop Rules in AI Coding Logs, all by A.G. Mauro and C.A. Harris, then stabilize bounded interpretations of that corpus: proof-carrying development, cached context as infrastructure, tool calls as the medium of work, prompt language as an operating system, and failures or stop rules as first-class evidence states. Portfolio-Scale Agentic Development Ecology: A Longitudinal Single-Operator Study is the capstone over that chain. It claims that, in this local archive, agentic development appears as a portfolio ecology whose recurring method is source-truth reopening, proof-aware change, evidence preservation, authority separation, and explicit closeout. It does not claim that the ecology is representative of other users, teams, organizations, model systems, or software portfolios. It does not claim productivity improvement, model quality, causal effect, benchmark superiority, public corpus release, or generalized organizational practice. The contribution is a final-local research frame for studying the portfolio as a system of interacting evidence practices while keeping every inherited claim below the authority already granted by the preceding A.G. Mauro and C.A. Harris corpus and interpretation manuscripts.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-20-portfolio-scale-agentic-development-ecology-a-longitudinal-single-operator-study", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-21-authority-engineering-language-a-machine-readable-contract-for-legitimate-action", "path": "/writing/papers/21-authority-engineering-language-a-machine-readable-contract-for-legitimate-action.html", "file": "writing/papers/21-authority-engineering-language-a-machine-readable-contract-for-legitimate-action.html", "url": "https://standframe.com/writing/papers/21-authority-engineering-language-a-machine-readable-contract-for-legitimate-action.html", "kind": "research-paper", "title": "Authority Engineering Language: A Machine-Readable Contract for Legitimate Action - Standing Framework", "description": "Agent systems need more than instructions that say \"ask before dangerous actions.\" They need machine-readable authority contracts: who may decide, which action is at stake, what evidence is required, which scope applies, what obligations constrain the decision, and where escalation goes when authority is missing. This paper describes the Authority Engineering Language, or AEL, as a candidate language for representing those contracts. The local AEL v0.3.5 specification defines syntax, static semantics, runtime action-event evaluation, optional pattern and temporal modules, deterministic audit requirements, and a reference intermediate representation. The adjacent Authority Engineering materials define a compact ontology around Authority Units, Decision Points, Obligation Units, Evidence Units, and Scope Boundaries. This draft grounds that language in the Switchboard scenario-packet work developed in Switchboard Harness DSL: Runtime Lab Scenario Packets as Governed Agent Work Contracts by A.G. Mauro and C.A. Harris. AEL should not replace the scenario packet or runtime. Instead, it can express the authority and evidence policy inside a scenario packet so a governed run can refuse, block, or escalate for inspectable reasons. The contribution is a final-local language and methods paper: legitimate action can be declared, compiled, evaluated, and audited as a first-class contract, while legal validity and institutional authority remain outside the language unless separately supplied and accepted.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-21-authority-engineering-language-a-machine-readable-contract-for-legitimate-action", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-22-switchboard-harness-dsl-runtime-lab-scenario-packets-as-governed-agent-work-contracts", "path": "/writing/papers/22-switchboard-harness-dsl-runtime-lab-scenario-packets-as-governed-agent-work-contracts.html", "file": "writing/papers/22-switchboard-harness-dsl-runtime-lab-scenario-packets-as-governed-agent-work-contracts.html", "url": "https://standframe.com/writing/papers/22-switchboard-harness-dsl-runtime-lab-scenario-packets-as-governed-agent-work-contracts.html", "kind": "research-paper", "title": "Switchboard Harness DSL: Runtime Lab Scenario Packets as Governed Agent Work Contracts - Standing Framework", "description": "Agent harnesses are often described as prompts, tools, runners, or evaluation suites. For governed agent systems, that description is too loose. A harness must also say what work object exists, which operator may act, what evidence must be produced, which approvals or budgets constrain action, which proof closes the loop, and what packet remains afterward. This paper develops the Switchboard harness DSL as a local methods design around that problem. The central object is the Runtime Lab scenario packet: a declarative contract containing Exchange seed state, operator roster, Line fixtures, adapter posture, workflow expectations, evidence rails, approval and budget policy, proof commands, and archive format. The source evidence is design and implementation-adjacent rather than compiler-complete: the Runtime Lab 2.0 review names scenario packets as the reintegration target, the Switchboard harness benchmark contract separates closed-track proof from open-track comparison, and the eval harness types expose cases, triggers, hard checks, rubric checks, traces, reports, and comparison summaries. The contribution is a bounded language design for declaring governed agent work before execution. The claim ceiling is final-local methods design: a scenario packet can make agent evaluation inspectable and replayable without turning every scenario into a public benchmark or every successful run into release authority.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-22-switchboard-harness-dsl-runtime-lab-scenario-packets-as-governed-agent-work-contracts", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-23-harness-search-under-governance-optimizing-agent-scaffolds-without-losing-proof-boundaries", "path": "/writing/papers/23-harness-search-under-governance-optimizing-agent-scaffolds-without-losing-proof-boundaries.html", "file": "writing/papers/23-harness-search-under-governance-optimizing-agent-scaffolds-without-losing-proof-boundaries.html", "url": "https://standframe.com/writing/papers/23-harness-search-under-governance-optimizing-agent-scaffolds-without-losing-proof-boundaries.html", "kind": "research-paper", "title": "Harness Search Under Governance: Optimizing Agent Scaffolds Without Losing Proof Boundaries - Standing Framework", "description": "Agent performance depends on more than model weights. Context, prompts, tool exposure, memory, retrieval, environment snapshots, retry policy, approval policy, and evidence capture can all change what an agent accomplishes. That makes harness search attractive: instead of treating the scaffold around an agent as fixed, a system can search over candidate scaffold changes. But harness search is dangerous when the search process can mutate the proof boundary. A candidate may improve a score by changing timeout policy, exposing a forbidden tool, weakening evidence requirements, reading development traces, or moving open-track evidence into closed-track claims. This paper proposes harness search under governance: a protocol for optimizing agent scaffolds while freezing the authority, proof, leakage, model, adapter, timeout, environment, and claim boundaries that make comparison meaningful. The operating object follows from Switchboard Harness DSL: Runtime Lab Scenario Packets as Governed Agent Work Contracts by A.G. Mauro and C.A. Harris: the search should run over declared scenario packets and candidate mutation records, not loose prompt experiments. The source evidence is local and design-bounded: Switchboard's Meta-Harness horizon retargets broad search into one native quality-improvement packet, the benchmark contract separates closed-track proof from open-track comparison, and the Terminal-Bench failure-forensics packet shows why opaque failures must be diagnosed before widening. The contribution is a control-plane method, not a leaderboard claim.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-23-harness-search-under-governance-optimizing-agent-scaffolds-without-losing-proof-boundaries", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-24-lean-acceptance-is-not-semantic-acceptance-lessons-from-a-formal-conjectures-internal-run", "path": "/writing/papers/24-lean-acceptance-is-not-semantic-acceptance-lessons-from-a-formal-conjectures-internal-run.html", "file": "writing/papers/24-lean-acceptance-is-not-semantic-acceptance-lessons-from-a-formal-conjectures-internal-run.html", "url": "https://standframe.com/writing/papers/24-lean-acceptance-is-not-semantic-acceptance-lessons-from-a-formal-conjectures-internal-run.html", "kind": "research-paper", "title": "Lean Acceptance Is Not Semantic Acceptance: Lessons From a Formal Conjectures Internal Run - Standing Framework", "description": "Formal verification strengthens agent evaluation by giving mechanical proof checks. It does not eliminate semantic interpretation. This paper studies a boundary exposed by an internal Switchboard-mediated Formal Conjectures run: a Lean kernel can accept a frozen target while the mathematical answer still fails semantic review. The internal run completed 200 counted Lean attempts over FC100OpenSet1 and FC100SolvedSet1, with 100 percent Switchboard evidence coverage and zero missing mediation records. Machine accounting produced 63 inclusive accepted rows, 26 new target proofs, 37 existing clean proofs, 137 depends_on_sorry rows, and three timeouts. Expert semantic review of the three OpenSet machine-accepted targets accepted one as a faithful frozen-target proof, rejected two as inadequate answer instantiations, and found zero novel open-problem discoveries. The contribution is a reporting pattern for formal benchmarks: kernel validity, frozen-target validity, semantic fidelity, prior-result status, and novelty must remain distinct.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-24-lean-acceptance-is-not-semantic-acceptance-lessons-from-a-formal-conjectures-internal-run", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-25-public-intelligence-products-from-source-unit-evidence-graphs", "path": "/writing/papers/25-public-intelligence-products-from-source-unit-evidence-graphs.html", "file": "writing/papers/25-public-intelligence-products-from-source-unit-evidence-graphs.html", "url": "https://standframe.com/writing/papers/25-public-intelligence-products-from-source-unit-evidence-graphs.html", "kind": "research-paper", "title": "Public Intelligence Products From Source-Unit Evidence Graphs - Standing Framework", "description": "This paper studies a source public-record corpus as a source-unit evidence graph that can produce reader-facing intelligence without exceeding the public record. From Public Records to Publication Claims: Evidence Promotion for Public-Source Research by A.G. Mauro and C.A. Harris covers claim promotion from machine-organized rows to publication candidates. This paper covers a different transformation: turning a large public source spine into issue briefs, concern maps, diligence queues, decision-risk registers, and public-safe talking points. The source evidence includes source-unit manifests, public evidence graph, public-process claim matrix, evidence posture brief, authority record, sample issue brief, artifact examples, and validation reports. The contribution is a method for public intelligence products that preserve source links, denominator counts, quote boundaries, nonclaims, terminal boundaries, and decision-point posture.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-25-public-intelligence-products-from-source-unit-evidence-graphs", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-26-signal-box-evidence-ladder-from-dogfood-packets-to-benchmark-style-review-corpora", "path": "/writing/papers/26-signal-box-evidence-ladder-from-dogfood-packets-to-benchmark-style-review-corpora.html", "file": "writing/papers/26-signal-box-evidence-ladder-from-dogfood-packets-to-benchmark-style-review-corpora.html", "url": "https://standframe.com/writing/papers/26-signal-box-evidence-ladder-from-dogfood-packets-to-benchmark-style-review-corpora.html", "kind": "research-paper", "title": "Signal Box Evidence Ladder: From Dogfood Packets to Benchmark-Style Review Corpora - Standing Framework", "description": "Evidence for agent products should not move by vibes. A dogfood run, an internal field case, an outside-host packet, and a benchmark-style corpus each answer different questions. This paper studies Signal Box as a staged evidence ladder for review-packet systems. The local Spinoffs research tree contains Level 1 dogfood evidence, Level 2 internal field evidence, Level 3 outside-context evidence, and a Level 4 benchmark-style scaffold with an error taxonomy. The counted source reports preserve denominators and nonclaims: Level 1 records 77 candidate/run cases with 59 accepted Level 1 cases and 18 reviewed failed-run cases; Level 2 records five accepted internal Switchboard field cases; Level 3 records three accepted outside-context cases while refusing product-market-fit, repeated-adoption, public-adoption, benchmark, and general review-time claims; Level 4 remains a draft scaffold requiring taxonomy and redacted corpus work before freezing. The contribution is a claim architecture for agent-produced review packets. Signal Box evidence becomes stronger only when the next rung's evidence exists; it does not become stronger because the story wants to move.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-26-signal-box-evidence-ladder-from-dogfood-packets-to-benchmark-style-review-corpora", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-27-plugin-runtime-trust-boundaries-for-governed-agent-systems", "path": "/writing/papers/27-plugin-runtime-trust-boundaries-for-governed-agent-systems.html", "file": "writing/papers/27-plugin-runtime-trust-boundaries-for-governed-agent-systems.html", "url": "https://standframe.com/writing/papers/27-plugin-runtime-trust-boundaries-for-governed-agent-systems.html", "kind": "research-paper", "title": "Plugin Runtime Trust Boundaries for Governed Agent Systems - Standing Framework", "description": "Tool-using agents are exposed to untrusted data, but plugin-enabled agent systems add another layer of risk: installable runtime surfaces. A plugin is not only a tool. It has package bytes, a manifest, worker entrypoints, UI launchers, secrets, host APIs, approval paths, compatibility rules, and rollout state. This paper studies Switchboard's plugin runtime trust boundaries as a governed-agent security architecture. The source threat-model refresh records enforced boundaries around reviewed artifact records, hosted package-byte agreement, rollout attestation, sandboxed iframe UI launchers, explicit secret references, and distinct-reviewer dual control for high-risk approvals. It also records open follow-through around immutable artifact distribution and broader fuzz coverage. The contribution is a trust-boundary model for plugin systems: artifact review, distribution, network, secret, UI, worker capability, approval, and compatibility boundaries must be represented separately and produce auditable denial evidence.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-27-plugin-runtime-trust-boundaries-for-governed-agent-systems", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-28-local-model-admission-as-governed-infrastructure", "path": "/writing/papers/28-local-model-admission-as-governed-infrastructure.html", "file": "writing/papers/28-local-model-admission-as-governed-infrastructure.html", "url": "https://standframe.com/writing/papers/28-local-model-admission-as-governed-infrastructure.html", "kind": "research-paper", "title": "Local Model Admission as Governed Infrastructure - Standing Framework", "description": "Local model infrastructure is often evaluated with simple performance questions: does the model answer, how fast is it, and how many tokens per second can it produce? For governed agent systems, those questions are too narrow. A local model route is not only a provider daemon. It is an admission surface with host fingerprint, provider record, model inventory, startup and warmup behavior, queue wait, load duration, time to first token, total duration, throughput, fallback, cancellation, refusal, and runtime visibility. This paper studies local-model admission as governed infrastructure using Switchboard's local-model benchmark contract. The contract states that the local-model control plane is no longer the open question; the remaining proof burden is whether managed local inference on Apple Silicon is fast, stable, and truthful enough to widen. The contribution is an admission protocol: a model/provider pair becomes useful evidence only when it passes through Switchboard's governed route path, archives enough context for later audit, and remains revocable when host, provider, model, or route conditions drift. The claim ceiling is final-local methods design. Local success, if later observed, would still be limited by host class, provider version, model inventory, route policy, and acceptance thresholds.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-28-local-model-admission-as-governed-infrastructure", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-29-graph-native-manuscript-development-claim-bundles-evidence-anchors-and-verification-reports", "path": "/writing/papers/29-graph-native-manuscript-development-claim-bundles-evidence-anchors-and-verification-reports.html", "file": "writing/papers/29-graph-native-manuscript-development-claim-bundles-evidence-anchors-and-verification-reports.html", "url": "https://standframe.com/writing/papers/29-graph-native-manuscript-development-claim-bundles-evidence-anchors-and-verification-reports.html", "kind": "research-paper", "title": "Graph-Native Manuscript Development: Claim Bundles, Evidence Anchors, and Verification Reports - Standing Framework", "description": "Agent-written manuscripts can become polished before they become trustworthy. Graph-native manuscript development offers a different object: sources, claims, sections, citations, draft fragments, review findings, and revisions represented as an evidence graph. This paper studies GraphOrchestra as an early local-first system for that object model. The inspected run attempted to produce a trustworthy literature review of graph-orchestrated manuscript development. Its artifacts include an outline plan, claim bundles, and a verification report. The run is valuable because it did not pass cleanly. The outline reports 146 artifacts, 74 document sections, and 800 code symbols, but no paper nodes and no notebook cells. Claim bundles repeatedly show no citations and narrow source support. The verification report marks the Introduction and Related Work sections as failed, reports warnings elsewhere, and records guardrail violations around evidence support and ambiguity. The contribution is therefore not an automated-manuscript success story. It is a methods and negative-evidence paper: graph-native manuscript systems are promising when they make unsupported claims, citation gaps, numeric mismatches, polarity contradictions, and unresolved ambiguities visible before prose is treated as final.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-29-graph-native-manuscript-development-claim-bundles-evidence-anchors-and-verification-reports", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-30-shipability-as-evidence-product-release-readiness-without-launch-overclaim", "path": "/writing/papers/30-shipability-as-evidence-product-release-readiness-without-launch-overclaim.html", "file": "writing/papers/30-shipability-as-evidence-product-release-readiness-without-launch-overclaim.html", "url": "https://standframe.com/writing/papers/30-shipability-as-evidence-product-release-readiness-without-launch-overclaim.html", "kind": "research-paper", "title": "Shipability as Evidence: Product Release Readiness Without Launch Overclaim - Standing Framework", "description": "Product teams often speak about \"shipping\" as if it were one state. In agentic software, that is too coarse. A repo can be technically green while publication is held. A launch post can exist while no public release has occurred. A trust brief can be strong while the supported deployment lane remains narrow. This paper studies shipability as an evidence object rather than a launch claim. The Switchboard release-content matrix centers one honest operating loop: one Controller, one default Operator, one real workspace, one concrete Line, and one visible review packet with artifact, verification, and Approval/Clearance state. The RUN6 release-gate supersession shows the same boundary at the proof layer: stale red command failures were cleared, technical release-gate posture became green, and publish remained held because a clean release boundary and explicit console approval were still required. The contribution is a product-systems method for release readiness. A product can be shipable in its honest lane when supported scope, proof, trust boundary, operator surface, and release narrative align. That still does not mean it has launched, been externally validated, reached market fit, or earned general deployment authority.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-30-shipability-as-evidence-product-release-readiness-without-launch-overclaim", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-31-agentic-temporal-compression-iteration-density-and-retrospective-distance-in-ai-assisted-software-development", "path": "/writing/papers/31-agentic-temporal-compression-iteration-density-and-retrospective-distance-in-ai-assisted-software-development.html", "file": "writing/papers/31-agentic-temporal-compression-iteration-density-and-retrospective-distance-in-ai-assisted-software-development.html", "url": "https://standframe.com/writing/papers/31-agentic-temporal-compression-iteration-density-and-retrospective-distance-in-ai-assisted-software-development.html", "kind": "research-paper", "title": "Agentic Temporal Compression: Iteration Density and Retrospective Distance in AI-Assisted Software Development - Standing Framework", "description": "AI-assisted software development can change the density of work without changing the clock. A developer can complete prompt-response-action loops, tool inspections, code patches, tests, reversals, and design decisions in a single day. In calendar time, little time has passed. In project time, the system may have crossed several meaningful states. This paper proposes a construct for that discrepancy: agentic temporal compression, the subjective experience that recent project history feels unusually distant because a fixed interval contains unusually dense meaningful change. The paper separates the proposed workflow variable from the proposed subjective outcome. Iteration density is the number of meaningful development cycles, model-mediated updates, decisions, experiments, or architecture revisions completed per unit of clock time. Agentic temporal compression is the retrospective experience that recent work feels older, farther away, or more historically remote than its calendar age suggests. The contribution is not evidence that the phenomenon is widespread or caused by large language models. The contribution is a testable framing that connects agent-assisted development to established work on event segmentation, prediction error, temporal context, episodic memory, and retrospective duration judgment. The central hypothesis is that dense agentic workflows increase prediction-error events and mental-model revisions. Those revisions create salient event boundaries. Event boundaries and retrievable changes then increase the remembered density of an interval, making recent project states feel temporally distant. The paper outlines a staged research program: qualitative interviews, diary and telemetry studies, within-subject controlled tasks, and longitudinal field observation during transitions into agentic workflows.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-31-agentic-temporal-compression-iteration-density-and-retrospective-distance-in-ai-assisted-software-development", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-32-operational-graphs-for-agentic-work-context-evidence-authority-and-verification", "path": "/writing/papers/32-operational-graphs-for-agentic-work-context-evidence-authority-and-verification.html", "file": "writing/papers/32-operational-graphs-for-agentic-work-context-evidence-authority-and-verification.html", "url": "https://standframe.com/writing/papers/32-operational-graphs-for-agentic-work-context-evidence-authority-and-verification.html", "kind": "research-paper", "title": "Operational Graphs for Agentic Work: Context, Evidence, Authority, and Verification - Standing Framework", "description": "Agentic work is often described as prompts, loops, workflows, or model sessions. Those terms are useful, but they do not explain how a reviewer later knows what context was used, which evidence supported an action, which authority boundary applied, or what verification attached to the result. This paper proposes the operational graph as a missing middle object for agentic systems. An operational graph is a small, typed, inspectable relation model over work objects: tasks, runs, artifacts, approvals, checks, memory records, evidence packets, source paths, and proof receipts. Its purpose is not to be a universal ontology or a decorative visualization. Its purpose is to make agent work traversable and auditable. The local evidence comes primarily from Switchboard's work and context graph surfaces, Caliper's commit-pinned repository graph and telemetry-only measurement contract, and adjacent Research Papers manuscripts on proof-carrying development, authority contracts, governed harnesses, and advisory metrics. The strongest supported claim is a methods claim: graph-shaped operating records can help separate context assembly, evidence support, authority gating, and verification state. The manuscript does not claim that graphs improve agent quality, that graph orchestration outperforms linear workflows, or that any local graph grants release, legal, publication, or deployment authority.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-32-operational-graphs-for-agentic-work-context-evidence-authority-and-verification", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-33-evidence-bound-harness-engineering-building-agent-systems-where-proof-travels", "path": "/writing/papers/33-evidence-bound-harness-engineering-building-agent-systems-where-proof-travels.html", "file": "writing/papers/33-evidence-bound-harness-engineering-building-agent-systems-where-proof-travels.html", "url": "https://standframe.com/writing/papers/33-evidence-bound-harness-engineering-building-agent-systems-where-proof-travels.html", "kind": "research-paper", "title": "Evidence-Bound Harness Engineering: Building Agent Systems Where Proof Travels - Standing Framework", "description": "Evidence-bound harness engineering is the discipline of designing agent harnesses so work can move quickly while proof, authority, and claim boundaries remain inspectable. That is the category this paper names. Ordinary harness engineering asks how to make an agent more capable: better context, better tools, better memory, better runners, better feedback. Evidence-bound harness engineering asks a second question: when the agent produces something, what is that result allowed to mean? That second question changes the harness. A harness is no longer just the prompt, the tool list, the memory layer, the test runner, or the eval suite. Those things still matter, but they are not enough. Once an agent is doing consequential work, the harness also has to preserve what work was attempted, what evidence was produced, what proof ran, what authority applied, what failed, what stayed in the denominator, and what can honestly be claimed afterward. Without that structure, a fluent completion note can start to sound like done work. A green check can start to sound like release approval. A benchmark row can start to sound like model truth. A receipt can start to sound like accepted evidence. An evidence-bound harness keeps those meanings from drifting. It lets agents move quickly, but it makes the proof travel with the work. That is what makes the category worth naming: not more ceremony around agents, but a way to keep speed from turning into proof debt.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-33-evidence-bound-harness-engineering-building-agent-systems-where-proof-travels", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-34-a-replication-corpus-for-human-codex-software-work", "path": "/writing/papers/34-a-replication-corpus-for-human-codex-software-work.html", "file": "writing/papers/34-a-replication-corpus-for-human-codex-software-work.html", "url": "https://standframe.com/writing/papers/34-a-replication-corpus-for-human-codex-software-work.html", "kind": "research-paper", "title": "A Replication Corpus for Human-Codex Software Work - Standing Framework", "description": "This paper applies the aggregate-only method from A Longitudinal Corpus of Human-Codex Software Work to a second single-account Codex JSONL history. The snapshot cutoff is 2026-07-25T19:32:16.892Z. The parser enumerates 13,277 session files: 13,268 included active-root files and 9 included archived-root files. Of those, 12,084 sessions contain usable token counters and 1,193 do not. The first observed event timestamp is 2026-02-10T15:14:13.129Z; the latest included event timestamp is 2026-07-25T19:32:16.748Z. Summing the last cumulative token counter per token-instrumented session yields 47,043,309,796 total tokens, including 46,861,245,371 input tokens, 44,925,306,624 cached input tokens, 180,514,025 output tokens, and 63,263,061 reasoning output tokens. The same parse counts 38,983 user messages, 225,294 assistant messages, 539,831 function calls, and 523,942 shell command calls. The contribution is a replication-style aggregate descriptor that supports Paper 14's method claims across a second private Codex-log corpus without turning either corpus into a public transcript release, billing record, productivity measure, model comparison, or population-general workflow claim.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-34-a-replication-corpus-for-human-codex-software-work", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-35-standing-framework-ai-work-that-holds-up", "path": "/writing/papers/35-standing-framework-ai-work-that-holds-up.html", "file": "writing/papers/35-standing-framework-ai-work-that-holds-up.html", "url": "https://standframe.com/writing/papers/35-standing-framework-ai-work-that-holds-up.html", "kind": "research-paper", "title": "Standing Framework: AI Work That Holds Up - Standing Framework", "description": "AI makes work easier to produce and harder to account for. A task can end with a fluent answer, a changed file, a dashboard, or a passing check, yet still leave the important question unresolved: can anyone safely decide what just happened? Standing Framework builds for that moment. Its software, research, and applied systems are organized around AI work that has to hold up after the run. The work should carry its intent, evidence, uncertainty, approval state, and limits with it, so a person can continue, reject, approve, or hold the result without reconstructing the whole conversation. A selected 24-episode review of company work from March through September 2026 supports a narrow claim: Standing Framework can be understood as an AI systems company for evidence-bearing, authority-aware work. Market acceptance, customer validation, adoption, willingness to pay, product-market fit, publication approval, public release, legal/privacy clearance, source-permission clearance, expert participation, and raw-log publication remain unproven.", "lifecycle_state": "local-draft", "claim_ceiling": "final-local-research-methods", "include_in_sitemap": true, "include_in_llms": false, "source_ids": [ "papers-35-standing-framework-ai-work-that-holds-up", "research-paper-count-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "global-proof-boundary-001" ] }, { "id": "route-writing-papers-index", "path": "/writing/papers/index.html", "file": "writing/papers/index.html", "url": "https://standframe.com/writing/papers/index.html", "kind": "paper-index", "title": "Research papers - Standing Framework", "description": "Browse research papers by title, author, date, and abstract.", "lifecycle_state": "local-draft", "claim_ceiling": "research-methods-index", "include_in_sitemap": true, "include_in_llms": true, "source_ids": [ "website-paper-index", "publication-status-language", "proof-boundary", "research-paper-count-001", "research-website-publication-state-001", "global-proof-boundary-001" ], "claim_ids": [ "research-paper-count-001", "research-website-publication-state-001", "global-proof-boundary-001" ] } ], "count": 58, "proof_posture": "local-static-proof-only", "nonclaims": [ "This static API bundle alone does not prove production freshness after later local changes.", "This static API bundle does not prove indexing, ranking, model preference, customer validation, adoption, product-market fit, publication acceptance, or external demand.", "This static API bundle does not expose authenticated, write, upload, webhook, handoff, commerce, MCP, A2A, or WebMCP capabilities." ] }