Skip to content

stark-ai-de/agent-skills / plugins/stark-ai-developer/skills/architecture-compass

architecture-compass

Verified

Set up ADR governance, audit architecture and drift, or plan and execute bounded refactors and code simplification. Use when work needs binding architecture decisions, library reuse, shared-code ownership, provider-to-local mapping, or architecture PR review.

FileKindSize
SKILL.mdskill md16.1 KB
agents/openai.yamlconfig692 B
assets/adr-example.guide.mdmarkdown1.5 KB
assets/adr-example.long.mdmarkdown2.3 KB
assets/adr-example.short.mdmarkdown1.1 KB
assets/adr-index-template.mdmarkdown1.3 KB
assets/adr-template.guide.mdmarkdown1.2 KB
assets/adr-template.long.mdmarkdown1.3 KB
assets/adr-template.short.mdmarkdown1.1 KB
assets/agent-instructions-template.mdmarkdown5.9 KB
assets/module-contract-review.mdmarkdown2.1 KB
assets/new-repo-adoption-plan-template.mdmarkdown6.7 KB
assets/openai-icon.pngbinary62.8 KB
assets/product-planning-profile.mdmarkdown3.7 KB
assets/product-planning-work-item.mdmarkdown2.6 KB
assets/refactor-report-template.mdmarkdown12.2 KB
assets/setup-report-template.mdmarkdown15.0 KB
assets/testing-outcome-receipt-template.mdmarkdown6.0 KB
assets/vitest4-testing-profile.mdmarkdown9.0 KB
references/ac-adr-001-route-architecture-compass-through-canonical-adr-triplets.guide.mdmarkdown3.7 KB
references/ac-adr-001-route-architecture-compass-through-canonical-adr-triplets.long.mdmarkdown4.8 KB
references/ac-adr-001-route-architecture-compass-through-canonical-adr-triplets.short.mdmarkdown1.8 KB
references/ac-adr-002-select-actions-resolve-authority-and-record-guardrail-adoption.guide.mdmarkdown4.0 KB
references/ac-adr-002-select-actions-resolve-authority-and-record-guardrail-adoption.long.mdmarkdown6.4 KB
references/ac-adr-002-select-actions-resolve-authority-and-record-guardrail-adoption.short.mdmarkdown2.0 KB
references/ac-adr-003-coordinate-agents-and-execute-only-approved-bounded-slices.guide.mdmarkdown8.5 KB
references/ac-adr-003-coordinate-agents-and-execute-only-approved-bounded-slices.long.mdmarkdown6.4 KB
references/ac-adr-003-coordinate-agents-and-execute-only-approved-bounded-slices.short.mdmarkdown1.9 KB
references/ac-adr-004-report-staged-evidence-and-protect-public-outputs.guide.mdmarkdown6.3 KB
references/ac-adr-004-report-staged-evidence-and-protect-public-outputs.long.mdmarkdown5.9 KB
references/ac-adr-004-report-staged-evidence-and-protect-public-outputs.short.mdmarkdown1.8 KB
references/ac-adr-005-make-repository-adrs-binding-agent-guardrails.guide.mdmarkdown5.1 KB
references/ac-adr-005-make-repository-adrs-binding-agent-guardrails.long.mdmarkdown5.5 KB
references/ac-adr-005-make-repository-adrs-binding-agent-guardrails.short.mdmarkdown1.6 KB
references/ac-adr-006-assign-workspace-ownership-and-source-roles.guide.mdmarkdown17.7 KB
references/ac-adr-006-assign-workspace-ownership-and-source-roles.long.mdmarkdown5.7 KB
references/ac-adr-006-assign-workspace-ownership-and-source-roles.short.mdmarkdown1.7 KB
references/ac-adr-007-enforce-runtime-safe-module-and-public-package-boundaries.guide.mdmarkdown4.3 KB
references/ac-adr-007-enforce-runtime-safe-module-and-public-package-boundaries.long.mdmarkdown5.7 KB
references/ac-adr-007-enforce-runtime-safe-module-and-public-package-boundaries.short.mdmarkdown1.8 KB
references/ac-adr-008-compose-nextjs-routes-rendering-and-component-responsibilities.guide.mdmarkdown6.5 KB
references/ac-adr-008-compose-nextjs-routes-rendering-and-component-responsibilities.long.mdmarkdown6.6 KB
references/ac-adr-008-compose-nextjs-routes-rendering-and-component-responsibilities.short.mdmarkdown1.8 KB
references/ac-adr-009-choose-read-query-caching-and-freshness-boundaries.guide.mdmarkdown14.1 KB
references/ac-adr-009-choose-read-query-caching-and-freshness-boundaries.long.mdmarkdown5.3 KB
references/ac-adr-009-choose-read-query-caching-and-freshness-boundaries.short.mdmarkdown2.5 KB
references/ac-adr-010-protect-writes-behind-validated-command-boundaries.guide.mdmarkdown6.2 KB
references/ac-adr-010-protect-writes-behind-validated-command-boundaries.long.mdmarkdown4.6 KB
references/ac-adr-010-protect-writes-behind-validated-command-boundaries.short.mdmarkdown2.6 KB
references/ac-adr-011-compose-long-running-backend-runtimes-and-lifecycles-explicitly.guide.mdmarkdown27.4 KB
references/ac-adr-011-compose-long-running-backend-runtimes-and-lifecycles-explicitly.long.mdmarkdown5.0 KB
references/ac-adr-011-compose-long-running-backend-runtimes-and-lifecycles-explicitly.short.mdmarkdown2.7 KB
references/ac-adr-012-resolve-environment-and-configuration-at-deployable-boundaries.guide.mdmarkdown4.6 KB
references/ac-adr-012-resolve-environment-and-configuration-at-deployable-boundaries.long.mdmarkdown4.8 KB
references/ac-adr-012-resolve-environment-and-configuration-at-deployable-boundaries.short.mdmarkdown2.2 KB
references/ac-adr-013-own-language-package-build-lint-and-supply-chain-tooling-explicitly.guide.mdmarkdown6.7 KB
references/ac-adr-013-own-language-package-build-lint-and-supply-chain-tooling-explicitly.long.mdmarkdown5.9 KB
references/ac-adr-013-own-language-package-build-lint-and-supply-chain-tooling-explicitly.short.mdmarkdown2.3 KB
references/ac-adr-014-select-application-runtimes-deployment-hosts-and-additional-targets-by-evidence.guide.mdmarkdown6.3 KB
references/ac-adr-014-select-application-runtimes-deployment-hosts-and-additional-targets-by-evidence.long.mdmarkdown5.4 KB
references/ac-adr-014-select-application-runtimes-deployment-hosts-and-additional-targets-by-evidence.short.mdmarkdown2.1 KB
references/ac-adr-015-select-frontend-capability-libraries-by-product-need.guide.mdmarkdown4.6 KB
references/ac-adr-015-select-frontend-capability-libraries-by-product-need.long.mdmarkdown5.5 KB
references/ac-adr-015-select-frontend-capability-libraries-by-product-need.short.mdmarkdown2.7 KB
references/ac-adr-016-select-ai-model-streaming-ui-and-agent-capabilities-deliberately.guide.mdmarkdown4.9 KB
references/ac-adr-016-select-ai-model-streaming-ui-and-agent-capabilities-deliberately.long.mdmarkdown6.0 KB
references/ac-adr-016-select-ai-model-streaming-ui-and-agent-capabilities-deliberately.short.mdmarkdown2.7 KB
references/ac-adr-017-select-relational-cache-queue-and-realtime-capabilities-by-data-requirements.guide.mdmarkdown6.7 KB
references/ac-adr-017-select-relational-cache-queue-and-realtime-capabilities-by-data-requirements.long.mdmarkdown5.6 KB
references/ac-adr-017-select-relational-cache-queue-and-realtime-capabilities-by-data-requirements.short.mdmarkdown2.8 KB
references/ac-adr-018-validate-behavior-at-the-owning-boundary-and-promote-enforcement-gradually.guide.mdmarkdown3.5 KB
references/ac-adr-018-validate-behavior-at-the-owning-boundary-and-promote-enforcement-gradually.long.mdmarkdown5.9 KB
references/ac-adr-018-validate-behavior-at-the-owning-boundary-and-promote-enforcement-gradually.short.mdmarkdown2.1 KB
references/ac-adr-019-apply-security-and-privacy-controls-at-every-trust-boundary.guide.mdmarkdown3.5 KB
references/ac-adr-019-apply-security-and-privacy-controls-at-every-trust-boundary.long.mdmarkdown6.1 KB
references/ac-adr-019-apply-security-and-privacy-controls-at-every-trust-boundary.short.mdmarkdown2.0 KB
references/ac-adr-020-define-data-ownership-tenancy-retention-and-deletion-before-access-paths.guide.mdmarkdown3.7 KB
references/ac-adr-020-define-data-ownership-tenancy-retention-and-deletion-before-access-paths.long.mdmarkdown5.0 KB
references/ac-adr-020-define-data-ownership-tenancy-retention-and-deletion-before-access-paths.short.mdmarkdown1.9 KB
references/ac-adr-021-preserve-compatibility-through-explicit-migrations-and-deprecation-windows.guide.mdmarkdown3.8 KB
references/ac-adr-021-preserve-compatibility-through-explicit-migrations-and-deprecation-windows.long.mdmarkdown5.6 KB
references/ac-adr-021-preserve-compatibility-through-explicit-migrations-and-deprecation-windows.short.mdmarkdown2.0 KB
references/ac-adr-022-deliver-reversible-slices-with-explicit-rollback-and-promotion-gates.guide.mdmarkdown3.4 KB
references/ac-adr-022-deliver-reversible-slices-with-explicit-rollback-and-promotion-gates.long.mdmarkdown5.0 KB
references/ac-adr-022-deliver-reversible-slices-with-explicit-rollback-and-promotion-gates.short.mdmarkdown1.9 KB
references/ac-adr-023-operate-services-with-observable-health-readiness-failure-and-cleanup.guide.mdmarkdown3.4 KB
references/ac-adr-023-operate-services-with-observable-health-readiness-failure-and-cleanup.long.mdmarkdown5.7 KB
references/ac-adr-023-operate-services-with-observable-health-readiness-failure-and-cleanup.short.mdmarkdown1.9 KB
references/ac-adr-024-meet-an-explicit-accessibility-baseline-with-automated-and-manual-proof.guide.mdmarkdown3.7 KB
references/ac-adr-024-meet-an-explicit-accessibility-baseline-with-automated-and-manual-proof.long.mdmarkdown5.7 KB
references/ac-adr-024-meet-an-explicit-accessibility-baseline-with-automated-and-manual-proof.short.mdmarkdown2.0 KB
references/ac-adr-025-set-measurable-performance-budgets-and-optimize-from-evidence.guide.mdmarkdown3.4 KB
references/ac-adr-025-set-measurable-performance-budgets-and-optimize-from-evidence.long.mdmarkdown5.7 KB
references/ac-adr-025-set-measurable-performance-budgets-and-optimize-from-evidence.short.mdmarkdown2.0 KB
references/ac-adr-026-route-architecture-compass-through-explicit-setup-and-apply-pipelines.guide.mdmarkdown6.0 KB
references/ac-adr-026-route-architecture-compass-through-explicit-setup-and-apply-pipelines.long.mdmarkdown10.0 KB
references/ac-adr-026-route-architecture-compass-through-explicit-setup-and-apply-pipelines.short.mdmarkdown2.5 KB
references/ac-adr-027-use-the-open-agent-skills-specification.guide.mdmarkdown2.0 KB
references/ac-adr-027-use-the-open-agent-skills-specification.long.mdmarkdown3.2 KB
references/ac-adr-027-use-the-open-agent-skills-specification.short.mdmarkdown1.6 KB
references/ac-adr-028-keep-candidates-outside-the-promoted-public-catalog.guide.mdmarkdown2.6 KB
references/ac-adr-028-keep-candidates-outside-the-promoted-public-catalog.long.mdmarkdown3.2 KB
references/ac-adr-028-keep-candidates-outside-the-promoted-public-catalog.short.mdmarkdown1.6 KB
references/ac-adr-029-promote-skills-through-an-evidence-and-maintenance-gate.guide.mdmarkdown2.4 KB
references/ac-adr-029-promote-skills-through-an-evidence-and-maintenance-gate.long.mdmarkdown3.5 KB
references/ac-adr-029-promote-skills-through-an-evidence-and-maintenance-gate.short.mdmarkdown1.6 KB
references/ac-adr-030-license-public-skill-repositories-under-apache-2-0.guide.mdmarkdown2.2 KB
references/ac-adr-030-license-public-skill-repositories-under-apache-2-0.long.mdmarkdown3.5 KB
references/ac-adr-030-license-public-skill-repositories-under-apache-2-0.short.mdmarkdown1.8 KB
references/ac-adr-031-keep-skill-evaluation-evidence-outside-the-runtime-payload.guide.mdmarkdown2.1 KB
references/ac-adr-031-keep-skill-evaluation-evidence-outside-the-runtime-payload.long.mdmarkdown3.6 KB
references/ac-adr-031-keep-skill-evaluation-evidence-outside-the-runtime-payload.short.mdmarkdown1.8 KB
references/ac-adr-032-keep-maintainer-local-helper-state-out-of-public-repositories.guide.mdmarkdown2.3 KB
references/ac-adr-032-keep-maintainer-local-helper-state-out-of-public-repositories.long.mdmarkdown3.7 KB
references/ac-adr-032-keep-maintainer-local-helper-state-out-of-public-repositories.short.mdmarkdown1.9 KB
references/ac-adr-033-choose-portable-dependency-light-skill-helpers.guide.mdmarkdown2.6 KB
references/ac-adr-033-choose-portable-dependency-light-skill-helpers.long.mdmarkdown3.9 KB
references/ac-adr-033-choose-portable-dependency-light-skill-helpers.short.mdmarkdown1.8 KB
references/ac-adr-034-keep-release-metadata-coherent-with-public-catalog-changes.guide.mdmarkdown2.3 KB
references/ac-adr-034-keep-release-metadata-coherent-with-public-catalog-changes.long.mdmarkdown3.8 KB
references/ac-adr-034-keep-release-metadata-coherent-with-public-catalog-changes.short.mdmarkdown1.8 KB
references/ac-adr-035-classify-skill-portability-before-choosing-host-variants.guide.mdmarkdown2.9 KB
references/ac-adr-035-classify-skill-portability-before-choosing-host-variants.long.mdmarkdown4.6 KB
references/ac-adr-035-classify-skill-portability-before-choosing-host-variants.short.mdmarkdown1.9 KB
references/ac-adr-036-keep-architecture-compass-portable-through-host-adapters.guide.mdmarkdown19.3 KB
references/ac-adr-036-keep-architecture-compass-portable-through-host-adapters.long.mdmarkdown4.2 KB
references/ac-adr-036-keep-architecture-compass-portable-through-host-adapters.short.mdmarkdown1.9 KB
references/ac-adr-037-preserve-target-contracts-and-gate-gateway-extraction.guide.mdmarkdown2.8 KB
references/ac-adr-037-preserve-target-contracts-and-gate-gateway-extraction.long.mdmarkdown4.3 KB
references/ac-adr-037-preserve-target-contracts-and-gate-gateway-extraction.short.mdmarkdown1.8 KB
references/ac-adr-038-gate-optional-capabilities-and-tool-side-effects.guide.mdmarkdown2.3 KB
references/ac-adr-038-gate-optional-capabilities-and-tool-side-effects.long.mdmarkdown4.0 KB
references/ac-adr-038-gate-optional-capabilities-and-tool-side-effects.short.mdmarkdown1.8 KB
references/ac-adr-039-prefer-existing-public-skills-conditionally.guide.mdmarkdown2.4 KB
references/ac-adr-039-prefer-existing-public-skills-conditionally.long.mdmarkdown4.3 KB
references/ac-adr-039-prefer-existing-public-skills-conditionally.short.mdmarkdown1.8 KB
references/ac-adr-040-offer-an-opinionated-stack-profile-for-new-public-skill-repositories.guide.mdmarkdown5.4 KB
references/ac-adr-040-offer-an-opinionated-stack-profile-for-new-public-skill-repositories.long.mdmarkdown5.5 KB
references/ac-adr-040-offer-an-opinionated-stack-profile-for-new-public-skill-repositories.short.mdmarkdown2.1 KB
references/ac-adr-041-publish-public-skills-through-github-and-the-open-skills-cli.guide.mdmarkdown2.7 KB
references/ac-adr-041-publish-public-skills-through-github-and-the-open-skills-cli.long.mdmarkdown4.2 KB
references/ac-adr-041-publish-public-skills-through-github-and-the-open-skills-cli.short.mdmarkdown1.9 KB
references/ac-adr-042-calibrate-validation-to-change-risk-and-reuse-fresh-evidence.guide.mdmarkdown7.9 KB
references/ac-adr-042-calibrate-validation-to-change-risk-and-reuse-fresh-evidence.long.mdmarkdown13.9 KB
references/ac-adr-042-calibrate-validation-to-change-risk-and-reuse-fresh-evidence.short.mdmarkdown3.2 KB
references/ac-adr-043-route-architecture-compass-through-explicit-setup-and-apply-pipelines-with-risk-based-validation.guide.mdmarkdown6.1 KB
references/ac-adr-043-route-architecture-compass-through-explicit-setup-and-apply-pipelines-with-risk-based-validation.long.mdmarkdown12.2 KB
references/ac-adr-043-route-architecture-compass-through-explicit-setup-and-apply-pipelines-with-risk-based-validation.short.mdmarkdown3.0 KB
references/ac-adr-044-record-material-decision-lineage-in-non-normative-guides.guide.mdmarkdown3.6 KB
references/ac-adr-044-record-material-decision-lineage-in-non-normative-guides.long.mdmarkdown6.3 KB
references/ac-adr-044-record-material-decision-lineage-in-non-normative-guides.short.mdmarkdown2.4 KB
references/ac-adr-045-route-architecture-compass-through-setup-audit-and-refactor-with-intent-bound-agent-selection.guide.mdmarkdown10.5 KB
references/ac-adr-045-route-architecture-compass-through-setup-audit-and-refactor-with-intent-bound-agent-selection.long.mdmarkdown6.4 KB
references/ac-adr-045-route-architecture-compass-through-setup-audit-and-refactor-with-intent-bound-agent-selection.short.mdmarkdown3.1 KB
references/ac-adr-046-rank-architecture-evidence-without-expanding-operational-authority.guide.mdmarkdown3.4 KB
references/ac-adr-046-rank-architecture-evidence-without-expanding-operational-authority.long.mdmarkdown5.8 KB
references/ac-adr-046-rank-architecture-evidence-without-expanding-operational-authority.short.mdmarkdown2.0 KB
references/ac-adr-047-separate-low-and-moderate-validation-risk-with-explicit-triggers.guide.mdmarkdown7.8 KB
references/ac-adr-047-separate-low-and-moderate-validation-risk-with-explicit-triggers.long.mdmarkdown13.7 KB
references/ac-adr-047-separate-low-and-moderate-validation-risk-with-explicit-triggers.short.mdmarkdown3.4 KB
references/ac-adr-048-persist-approved-governance-before-planned-architecture-refactors.guide.mdmarkdown10.6 KB
references/ac-adr-048-persist-approved-governance-before-planned-architecture-refactors.long.mdmarkdown8.7 KB
references/ac-adr-048-persist-approved-governance-before-planned-architecture-refactors.short.mdmarkdown3.1 KB
references/ac-adr-049-distinguish-change-risk-from-representative-environment-observation.guide.mdmarkdown9.9 KB
references/ac-adr-049-distinguish-change-risk-from-representative-environment-observation.long.mdmarkdown16.3 KB
references/ac-adr-049-distinguish-change-risk-from-representative-environment-observation.short.mdmarkdown4.1 KB
references/ac-adr-050-use-semantic-status-markers-in-user-facing-receipts.guide.mdmarkdown3.3 KB
references/ac-adr-050-use-semantic-status-markers-in-user-facing-receipts.long.mdmarkdown4.8 KB
references/ac-adr-050-use-semantic-status-markers-in-user-facing-receipts.short.mdmarkdown1.8 KB
references/ac-adr-051-route-architecture-compass-through-public-and-internal-decision-namespaces.guide.mdmarkdown3.9 KB
references/ac-adr-051-route-architecture-compass-through-public-and-internal-decision-namespaces.long.mdmarkdown7.4 KB
references/ac-adr-051-route-architecture-compass-through-public-and-internal-decision-namespaces.short.mdmarkdown2.3 KB
references/ac-adr-052-persist-agent-governance-through-host-neutral-repository-surfaces.guide.mdmarkdown4.4 KB
references/ac-adr-052-persist-agent-governance-through-host-neutral-repository-surfaces.long.mdmarkdown7.0 KB
references/ac-adr-052-persist-agent-governance-through-host-neutral-repository-surfaces.short.mdmarkdown2.1 KB
references/ac-adr-053-use-capability-aware-presentation-profiles-for-portable-agent-receipts.guide.mdmarkdown5.0 KB
references/ac-adr-053-use-capability-aware-presentation-profiles-for-portable-agent-receipts.long.mdmarkdown8.5 KB
references/ac-adr-053-use-capability-aware-presentation-profiles-for-portable-agent-receipts.short.mdmarkdown2.2 KB
references/ac-adr-054-isolate-agent-writes-in-external-git-worktrees.guide.mdmarkdown11.5 KB
references/ac-adr-054-isolate-agent-writes-in-external-git-worktrees.long.mdmarkdown7.0 KB
references/ac-adr-054-isolate-agent-writes-in-external-git-worktrees.short.mdmarkdown3.5 KB
references/ac-adr-055-separate-change-impact-from-generated-release-publication.guide.mdmarkdown4.3 KB
references/ac-adr-055-separate-change-impact-from-generated-release-publication.long.mdmarkdown5.4 KB
references/ac-adr-055-separate-change-impact-from-generated-release-publication.short.mdmarkdown2.3 KB
references/ac-adr-056-preserve-release-candidates-through-verified-protected-history.guide.mdmarkdown2.4 KB
references/ac-adr-056-preserve-release-candidates-through-verified-protected-history.long.mdmarkdown3.2 KB
references/ac-adr-056-preserve-release-candidates-through-verified-protected-history.short.mdmarkdown1.8 KB
references/ac-adr-057-separate-metadata-repair-from-installable-artifact-provenance.guide.mdmarkdown2.7 KB
references/ac-adr-057-separate-metadata-repair-from-installable-artifact-provenance.long.mdmarkdown3.6 KB
references/ac-adr-057-separate-metadata-repair-from-installable-artifact-provenance.short.mdmarkdown1.9 KB
references/ac-adr-058-use-pnpm-for-package-management-and-bun-for-execution.guide.mdmarkdown7.1 KB
references/ac-adr-058-use-pnpm-for-package-management-and-bun-for-execution.long.mdmarkdown8.0 KB
references/ac-adr-058-use-pnpm-for-package-management-and-bun-for-execution.short.mdmarkdown3.4 KB
references/ac-adr-059-own-repository-validation-through-discoverable-framework-tests.guide.mdmarkdown2.8 KB
references/ac-adr-059-own-repository-validation-through-discoverable-framework-tests.long.mdmarkdown5.3 KB
references/ac-adr-059-own-repository-validation-through-discoverable-framework-tests.short.mdmarkdown1.8 KB
references/ac-adr-060-isolate-test-execution-by-effects-and-runtime-contracts.guide.mdmarkdown2.4 KB
references/ac-adr-060-isolate-test-execution-by-effects-and-runtime-contracts.long.mdmarkdown5.2 KB
references/ac-adr-060-isolate-test-execution-by-effects-and-runtime-contracts.short.mdmarkdown1.6 KB
references/ac-adr-061-shard-tests-as-complete-fail-closed-evidence-sets.guide.mdmarkdown2.7 KB

assets/refactor-report-template.md

View on GitHub
# Architecture Audit and Refactor Report

> Derived, non-normative asset. The applicable canonical Long ADRs prevail if this template conflicts or drifts. Include only applicable sections; keep activation compact and report each fact once.

## Intent-bound selection

- Public workflows exposed: `setup | audit | refactor | plan-refactor | plan-run-refactor`
- Selected workflow: `audit | refactor | plan-refactor | plan-run-refactor`
- Selection rationale and intent evidence:
- Write scope: `read-only | approved specification/ADR paths | accepted-ADR-governed implementation paths`
- Planning capability: `<state - evidence>`
- Read-only enforcement: `<state - evidence>`
- Architecture decision status: `<not required | pending | approved | blocked>`
- Execution status: `<not requested | ready for direct execution | pending Plan-mode exit | pending write permission | blocked | completed>`
- Expected artifacts:
- Protected paths and pre-existing state:
- Separate approval boundaries:

## Inspected evidence

| Evidence                                | Status | Revision/path | Notes |
| --------------------------------------- | ------ | ------------- | ----- |
| Repository identity and protected state |        |               |       |
| Agent instructions                      |        |               |       |
| ADR catalog, local index, and mapping   |        |               |       |
| Stack/architecture docs                 |        |               |       |
| Representative code/tests               |        |               |       |
| Validation/CI                           |        |               |       |
| Repository-native validation receipts   |        |               |       |
| Stable public-skill workflow surface    |        |               |       |

## Selected decisions

Use AC-ADR-046 strengths without confusing architecture authority with execution permission: `required` is an applicable accepted local ADR or mandatory target rule; `preferred` is a target-documented default with an allowed evidence-backed deviation; `example` is approved shape guidance; `assumption` is unverified and cannot justify mutation. Lower-ranked current code, provider decisions, and framework defaults cannot override a contradictory `required` rule.

| Rule | Rule strength | Provider ADR Short | Canonical Long | Local ADR/path | Target provenance | Applies to |
| ---- | ------------- | ------------------ | -------------- | -------------- | ----------------- | ---------- |
|      |               |                    |                |                |                   |            |

## Adopted Jev host advice

- Provider candidate: `AC-ADR-065`; disposition and target evidence:
- Accepted repository-native ADR, provider mapping, and applicable repository scope:
- Authorized hosts and configuration owner:
- Missing prerequisites, unmet obligation, native fallback, owner, and revisit trigger:

| Host and repository scope | Installed artifact | Configured registration | Effectively active/trusted | Explicit processing authority and approved metadata scope | Current inventory, restrictions, and coverage limits | Qualified host/runtime/configuration/model/reasoning evidence | Evidence status, source, observed at, and invalidators |
| ------------------------- | ------------------ | ----------------------- | -------------------------- | --------------------------------------------------------- | ---------------------------------------------------- | ------------------------------------------------------------- | ------------------------------------------------------ |
|                           |                    |                         |                            |                                                           |                                                      |                                                               |                                                        |

Use separate evidence for each dimension; adoption is not activation, processing consent, current eligibility, or qualification. Mark unknown, missing, failed, or stale evidence truthfully. Preserve explicit user skill choices and native selection when prerequisites are unmet. Audit reports existing evidence only: no provider requests, qualification probes, repairs, or writes. Setup records governance and gaps without changing global host configuration.

## Findings and disposition

Assign each finding exactly one disposition: `fix-and-prove-now`, `verify-now`, `defer-recorded`, `accept-risk`, or `not-applicable`.

Severity definitions:

- `critical`: active or imminent loss of confidentiality, integrity, availability, data, or control with no safe bounded continuation.
- `high`: material violation of a binding decision, trust boundary, public contract, migration safety, or release gate that blocks the affected work.
- `moderate`: behavioral or governance drift with bounded impact and a practical workaround, but which does not meet the critical/high triggers.
- `low`: localized non-behavioral inconsistency, clarity defect, or maintenance risk with deterministic acceptance and no higher-severity trigger.

| Severity | File/area | Concrete drift or risk | Governing ADR | Rule strength | Disposition | Recommended Action | Docs/ADR Impact | Deviation Resolution | Proof obligation | Done-When |
| -------- | --------- | ---------------------- | ------------- | ------------- | ----------- | ------------------ | --------------- | -------------------- | ---------------- | --------- |
|          |           |                        |               |               |             |                    |                 |                      |                  |           |

If two applicable sources conflict, do not blend them. Record both sources, operational authority, affected scope and impact, recommended resolution, decision owner, and whether disjoint work can continue; stop the dependent slice until resolved.

## Workflow boundary

- Audit remained strictly read-only:
- Direct refactor was already fully governed and bounded:
- Missing governance routed to `setup` rather than repaired silently:
- Unresolved durable decisions routed to a Plan workflow:
- Material reclassification and authority resolution:

## Conditional stable-skill selector instruction

- Applicability: `applicable | not applicable | indeterminate`
- Target evidence:
- Existing equivalent or conflicting rule:
- Handling: `audit report only | setup required | preserved | not applicable | indeterminate`

## Approved refactoring specification

Complete this section for `plan-refactor` and `plan-run-refactor`.

- Native or conversational planning and material host-state evidence:
- Approval revision, exact scope, and existing authority reused:
- Delivery: `persist | chat-only`
- Actual persistence status and paths (separate from approval):
- Approved outcome and non-goals:
- Durable decisions resolved:
- Exact path allowlist:
- Ordered reversible slices:
- Proof obligations, owners, and reusable receipts:
- Rollback and stop conditions:
- Separate external/high-risk approvals:
- Plan-mode exit evidence before persistence or execution, when required:
- Target-state recheck and any affected change requiring renewed approval:

## State recheck and execution

Complete this section for `refactor` and the execution phase of `plan-run-refactor`.

- Root, HEAD, index/worktree, dependency, permission, protected-path, and external-state recheck:
- Drift result: `unchanged | material drift - stopped`
- Implemented slices and exact paths:
- Integrated diff review:
- Deferred work:

For adopted testing decisions, link the target-native measurement record using [testing-outcome-receipt-template.md](testing-outcome-receipt-template.md). Preserve provider revision/content identity and local mapping there; do not create a second ledger. Measurement states (`met`, `unmet`, `unmeasured`, `waived`, `not-applicable`) do not replace execution or evidence statuses.

## Simplification evidence (when requested)

Use the optional [AC-ADR-006 loop](../references/ac-adr-006-assign-workspace-ownership-and-source-roles.guide.md#simplification-loop) only for a requested simplification assessment or implementation. Leave this section out for unrelated work; audit reports observations without persisting an artifact.

- Baseline / final revision or content identity; task delta versus full PR:
- Scope / tracked and untracked inventory / structural scan / semantic review / exclusions and unreadable paths:
- Applied, retained and deferred candidates; alternative APIs/versions and equivalence evidence:
- Shared-code owner, real consumers, runtime boundaries and remaining custom policy:
- Dependency changes, integration/upgrade responsibility and exit path:
- Counting method, paths, rename treatment, untracked additions, completeness and binary/generated exclusions:

| Category                                   | Added lines | Deleted lines | Net delta (added - deleted) |
| ------------------------------------------ | ----------- | ------------- | --------------------------- |
| Implementation                             |             |               |                             |
| Tests / fixtures                           |             |               |                             |
| Other (name docs, config, generated, etc.) |             |               |                             |
| All changed files                          |             |               |                             |

Keep the initial baseline fixed; identify each pass's starting candidate and count all added support files, including untracked additions without staging. Record a baseline bridge for concurrent edits or segmented work; do not claim a verified net reduction from incomplete totals. Separate handwritten production, tests/fixtures, docs/config, and generated/vendor/lockfile changes. Use the same formatter/counter and paths for incremental and cumulative measurements.

| Pass / candidate | Removed responsibility or replacement API/version | Handwritten delta / cumulative delta | Generated/vendor/lockfile delta | Existing proof receipt IDs |
| ---------------- | ------------------------------------------------- | ------------------------------------ | ------------------------------- | -------------------------- |
|                  |                                                   |                                      |                                 |                            |

- Proof: link the existing validation ledger below; keep unmeasured outcomes explicit.
- Final complete-pass coverage, stop reason, remaining candidates / rejection reasons and reopen triggers (distinguish convergence from blocked or incomplete work):

## Validation ledger

Record one AC-ADR-049 receipt per distinct proof obligation and reconcile every delegated receipt against the integrated candidate.

```text
Receipt ID:
Risk: low | moderate | high | critical
Cadence: reuse | final-batch | checkpointed | reproduce-first
Proof obligation:
Subject / owning boundary:
Revision / exact candidate artifact / dirty-tree or content fingerprint:
Command / scenario / harness:
Toolchain / config / fixtures / lockfile:
Evidence stage: source/static | local | CI | publication/install | deployed/production | external/third-party
Environment:
Status: verified | failed | not run | unavailable | stale
Observation / result:
Observed at / freshness boundary:
Covered contracts:
Invalidators / limitations:
Owner / source / run link:
Repository-native receipt location:
Skip reason: none | <why this obligation was not executed and who authorized it>
Final aggregate gate: <command or scenario, relationship to this receipt, and result>
```

| Receipt ID | Risk | Cadence | Proof obligation | Subject | Command/scenario/harness | Toolchain | Stage | Environment | Status | Observation/result | Invalidators | Owner | Skip reason | Final aggregate gate relationship |
| ---------- | ---- | ------- | ---------------- | ------- | ------------------------ | --------- | ----- | ----------- | ------ | ------------------ | ------------ | ----- | ----------- | --------------------------------- |
|            |      |         |                  |         |                          |           |       |             |        |                    |              |       |             |                                   |

## Remaining risk and next authorized action

-
rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first.Stars: 194kUpdated Claude CodeCodexVerifiedInstall