Repository navigation
AUDIT: executive three-product audit (Software Factory + StashMind + EasyTutor) - #7
Merged
Merged
Conversation
Why: Executive merge job Stage M0 verified both repos before building, and M1 wrote eight cross-linked architecture decision records. Per the Constitution, run state, M0 report, scorecard, and re-audit output live in runs/<id>/. What: Added runs/m0-20261021-102705/ with STATE.md, M0-REPORT.md, SCORECARD.md, and RE-AUDIT.md. Evidence: StashMind typecheck 0 errors, 43 tests pass, tag v1.0.0-rc1 present; EasyTutor typecheck 0 errors, lint 0 errors, 181 tests pass, tag v1.0.1 present; curl smoke tests pass on both live hosts. StashMind doctor: 69 PASS / 12 WARN / 10 FAIL (S0/S1 findings recorded). Doc drift fixed in both repos' READMEs (committed on their merge/stashmind-integration branches, PRs open, not merged). Tests: added 0 (M0/M1 are verification + architecture, no feature tests). Gates: StashMind verify green; EasyTutor typecheck/lint/test green. Merge-Layer: M0+M1 Counterpart: stashmind@d08e19e, easytutor@77d717a
Why: The StashMind x EasyTutor merge job (M0+M1 complete) needs a formal executive audit spanning Software Factory, StashMind, and EasyTutor before any further build layers run. The task document follows the strict format and decomposes into 8 independently verifiable units. What: Added tasks/executive-audit-three-product.md with milestones A1-A8 covering repo snapshots, live smoke tests, M0 re-audit, and the executive report publication. Evidence: task document parses cleanly; decomposes into 8 units, each within the 5-criterion cap, with verify commands and machine-checkable acceptance criteria. Tests: added 0 (task document only). Gates: npm run verify green (typecheck + test + build). Merge-Layer: audit-task
Why: The StashMind x EasyTutor merge job (M0+M1 complete) required a formal executive audit spanning Software Factory, StashMind, and EasyTutor. These artifacts record where each product stands, the plan and strategy forward, launch readiness and real-world usability, and the real-time work-management view. What: Added runs/audit-three-product/ with EXECUTIVE-AUDIT.md, PLAN-FORWARD.md, LAUNCH-READINESS.md, and REALTIME-VIEW.md. Every claim is backed by an executed command or committed artifact; no secrets are included. Evidence: task document parses into 8 units (within 5-criterion cap); npm run verify:loop 7/7; capability manifest 14 live / 4 blocked; StashMind gates green (43 tests, tag v1.0.0-rc1); EasyTutor gates green (181 tests, tag v1.0.1); live smoke tests 200/401/401/200. Tests: added 0 (audit artifacts only). Gates: npm run verify green. Merge-Layer: audit-report Counterpart: stashmind@eafa7f3, easytutor@51fb818
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
…anifest (Layer 1 Control) Add ADR-021 defining the seven-layer architecture (Control, Agency, Runtime, Adapter, Diagnostic, Observability, Interface) with fail-closed, no-second-writer, advisory-diagnostic, and dual-vocabulary invariants. Register preview-plane, bug-detective, dual-language-reporting, and tool-agnostic-interface in the capability manifest (dual-language-reporting live with proof; the rest planned, awaiting their layers). Add src/utils/dualLanguageReport.ts rendering every durable record in technical and founder/CEO vocabularies, with a named reproducing command for every figure. AI-Assisted: yes
Adds previewPlaneService managing per-tenant preview sessions: worktree creation, dev server start with port detection, session list/get/stop/ stopAll. Registers POST/GET/DELETE /api/preview/sessions routes under loop:read RBAC. Includes verify-preview-plane.sh gate and 6 tests. AI-Assisted: yes
… 5 Diagnostic) Adds the AI bug detective: error flagging, root-cause hypothesis, file:line pointers, remediation advice. Advisory only — never writes to a target tree; any fix still runs VERIFY -> REVIEW -> COMMIT through L2 gates. - Adds 'defect' failure class to loopFailureClass.ts with reason patterns (DEFECT_*, BUG_*) and gate mapping - Adds 'bug-diagnosis-evidence' gate id (advisory, never refuses) - bugDetectiveService: deterministic diagnosis from failure input - POST /api/bug-detective/diagnose route (loop:read RBAC) - API error codes BUG_DIAGNOSIS_NOT_FOUND, BUG_DIAGNOSIS_READ_ONLY - capability manifest: bug-detective and preview-plane moved to live - verify-bug-detective.sh gate script, 7 tests, all pass AI-Assisted: yes
Adds verify-observability.sh confirming read-only observability endpoints surface real durable state and blocked capabilities (observability-dashboards, cycle-time-metrics) name their operators. L6 is read-only views of durable records — dashboards, paging, and on-call routing remain blocked and operator-owned. AI-Assisted: yes
Updates README Architecture section from four boundaries to seven layers (L1 Control through L7 Interface) with invariants, and the capabilities table (regenerated: 17 live · 12 blocked · 1 planned) now includes preview-plane, bug-detective, and dual-language-reporting as live. CHANGELOG records Layer 5 bug detective (defect class, bug-diagnosis-evidence gate), Layer 6 observability gate, and Layer 2 preview plane promotion. Codebase index re-rendered. AI-Assisted: yes
This branch was successfully deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Executive audit of the StashMind x EasyTutor merge job across all three products.
M0 verification + M1 architecture complete. M2+ blocked on human gates H2/H3/H4/H6 and S0 credential fixes.
Reports: runs/audit-three-product/ (EXECUTIVE-AUDIT, PLAN-FORWARD, LAUNCH-READINESS, REALTIME-VIEW)
Task document: tasks/executive-audit-three-product.md (8 units, loop-verified)
Gates: npm run verify green; npm run verify:loop 7/7.
DO NOT MERGE — pending your decision on whether to keep PRs unmerged per the merge job's Definition of Done #6.