chore(delivery): align BT6 build and release gates (#160)

Make the contributor, CI, package, and tagged-release contracts agree with the shared BT6 baseline while preserving T3MP3ST's project-specific integrity checks. Exclude workspace-only material from published packages and keep local-agent detection tests host-independent.\n\nCloses #159
2026-08-23 16:22:24 -04:00
1 parent 6d4e01756a
commit c352f34843
@@ -0,0 +1,36 @@
+## Summary
+
+<!-- What changed and why? -->
+
+## Linked issue
+
+Closes #
+
+## Contribution receipt
+
+- Scope class: docs_only | static_fixture | local_lab | ctf_range | authorized_live
+- Target authority:
+- Network use: none | loopback | private_lab | authorized_external
+- Claims or evidence changed:
+- Redaction/provenance notes:
+
+## Verification
+
+- [ ] `npm run typecheck`
+- [ ] `npm test`
+- [ ] `npm run doctor`
+- [ ] Risk-specific checks are listed below, or not applicable with a reason
+
+Exact commands and results:
+
+## Risk and rollback
+
+- Residual risk:
+- Rollback:
+
+## Delivery checks
+
+- [ ] Diff is scoped against current `upstream/main`
+- [ ] No unrelated protected evidence, safety tests, or provider/config files were removed
+- [ ] Published review history was not force-pushed
+- [ ] Exact-head CI must be green before merge
@@ -2,13 +2,24 @@ name: T3MP3ST CI

 on:
  pull_request:
+    branches: [main]
  push:
    branches:
      - main
+    tags:
+      - "v*"
+
+permissions:
+  contents: read
+
+concurrency:
+  group: t3mp3st-ci-${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: ${{ github.event_name == 'pull_request' }}

 jobs:
  test:
    runs-on: ubuntu-latest
+    timeout-minutes: 20
    steps:
      - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4.3.1
      - uses: actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020 # v4.4.0
@@ -35,3 +46,35 @@ jobs:
      # Non-network smoke: core traps run deterministically; server/live tiers
      # auto-skip when no server/LLM is present (they are absent in CI).
      - run: npm run smoke
+
+  release:
+    if: startsWith(github.ref, 'refs/tags/v')
+    needs: test
+    runs-on: ubuntu-latest
+    timeout-minutes: 30
+    steps:
+      - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4.3.1
+        with:
+          ref: ${{ github.sha }}
+      - uses: actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020 # v4.4.0
+        with:
+          node-version: 22
+          cache: npm
+      - run: npm ci
+      - run: npm run test:release
+      - run: npm audit --audit-level=high
+      - run: npm run pack:dry-run
+      - run: npm pack --json > pack-result.json
+      - name: Record release checksums
+        run: |
+          package_file="$(node -e "const fs=require('fs'); const p=JSON.parse(fs.readFileSync('pack-result.json')); process.stdout.write(p[0].filename)")"
+          sha256sum "$package_file" > SHA256SUMS
+      - uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1
+        with:
+          name: tested-package-${{ github.sha }}
+          path: |
+            *.tgz
+            SHA256SUMS
+            pack-result.json
+          if-no-files-found: error
+          retention-days: 14
@@ -4,6 +4,12 @@ T3MP3ST needs contributions from prompt engineers, cyber operators, bug bounty h

 The best contributions make the system more capable while making its evidence and authority boundaries clearer.

+The canonical repository and tracker are
+[`elder-plinius/T3MP3ST`](https://github.com/elder-plinius/T3MP3ST). Open issues
+and pull requests there against `main`. The project uses focused branches,
+squash merges, green exact-head CI, and does not permit force-pushing published
+review history.
+
 ## High-Value Contribution Types

 - Add a tool adapter in `src/arsenal/catalog.ts`.
@@ -41,6 +47,12 @@ Prompt packs should include:

 ## Review Standard

+Behavior changes must include outcome-oriented regression coverage. A change
+with no relevant tests is blocked. Changed executable lines should reach at
+least 50% coverage; documentation-only and metadata-only changes may mark this
+not applicable with a reason. Trust-boundary, scope, evidence, provider,
+installation, and release changes also need the focused checks for that risk.
+
 Before opening a PR:

 ```bash
@@ -72,6 +84,14 @@ unrelated provider/config churn, benchmark fixture removals, provenance doc
 removals, or safety-test removals. If the branch has drifted, recreate it from
 current `main` and reapply only the intended change.

+Maintainers run `npm run test:release`, `npm audit --audit-level=high`, and the
+package dry run against the exact release commit before publishing. A green PR
+gate is necessary for review and merge, but it is not release certification.
+
+Do not include secrets, private tracker content, unlicensed corpora, or
+uncoordinated vulnerability details. Use the disclosure channel in
+`SECURITY.md` for security-sensitive reports.
+
 ## Style

 - Prefer clear adapters and evidence contracts over clever hidden behavior.
@@ -72,6 +72,14 @@ both human contributors and coding agents.
   - screenshots or artifacts for UI/reporting changes
   - residual risk or follow-up work

+7. Preserve review identity and history.
+
+   Sign contributor commits with a key associated with your forge identity (or
+   GitHub's verified web flow). Do not force-push a published review branch;
+   add corrective commits. Maintainers recheck the exact head, required CI,
+   mergeability, and linked issues immediately before the configured squash
+   merge.
+
 ## Coding-Agent Instructions

 When an agent prepares a PR, it must perform a final scope audit before posting.
@@ -139,3 +147,10 @@ feature details while stale-base deletions remain in the diff.
 Only comment on PRs that need action: concrete fixes, missing verification,
 missing receipts, or a rebase/scope cleanup. Ready PRs should not receive
 process comments.
+
+When a contributor has substantially completed a significant reviewed change
+and only narrow mechanical cleanup remains, maintainers may resolve the small
+conflict, preserve an established default, or add focused regression coverage
+as a courtesy. Re-run the full exact-head review and tests and tell the
+contributor what changed. Large rebases, behavior changes, architectural
+judgment, or newly discovered substantive defects remain contributor work.
@@ -19,6 +19,7 @@ npm run test:no-self-fitting
 npm run test:no-phantom-tools
 npm run test:gate
 npm run prompt:audit
+npm run pack:dry-run        # inspect the allowlisted package manifest; no workspace/private files
 ```

 ## 2. Local API smoke (loopback only)
@@ -56,8 +57,22 @@ degrade gracefully — they never fail the core run.

 ## 6. Publish

+- Confirm the release commit is already on `main`, CI is green at that exact
+  SHA, the version and changelog agree, and the tag is signed and points to that
+  commit. Never move or reuse a failed release tag; correct the release and use
+  a new version/tag.
 - Verify `repository.url` in `package.json` points at this repo. To retarget it:
  ```bash
  npm pkg set repository.url="git+https://github.com/<owner>/<repo>.git"
  ```
 - Tag the release only after Sections 1–2 are green.
+- Push the `v*` tag and wait for the tag workflow. It reruns
+  `npm run test:release`, the high-severity dependency audit, and the package
+  dry run against the exact tag, then retains the tested `.tgz`, its manifest,
+  and `SHA256SUMS` as workflow artifacts.
+- Inspect `pack-result.json`: only the allowlisted runtime, scripts, tools,
+  documentation, examples, and package metadata may ship. Workspace notes,
+  tests, AIWG/provider deployment internals, secrets, and local artifacts are
+  release blockers.
+- Publish the retained, checksum-matched package artifact. Do not rebuild a
+  different archive from another checkout after certification.
@@ -83,6 +83,14 @@ both human contributors and coding agents.
   - screenshots or artifacts for UI/reporting changes
   - residual risk or follow-up work

+7. Preserve review identity and history.
+
+   Sign contributor commits with a key associated with your forge identity (or
+   GitHub's verified web flow). Do not force-push a published review branch;
+   add corrective commits. Maintainers recheck the exact head, required CI,
+   mergeability, and linked issues immediately before the configured squash
+   merge.
+
 ## Coding-Agent Instructions

 When an agent prepares a PR, it must perform a final scope audit before posting.
@@ -150,3 +158,10 @@ feature details while stale-base deletions remain in the diff.
 Only comment on PRs that need action: concrete fixes, missing verification,
 missing receipts, or a rebase/scope cleanup. Ready PRs should not receive
 process comments.
+
+When a contributor has substantially completed a significant reviewed change
+and only narrow mechanical cleanup remains, maintainers may resolve the small
+conflict, preserve an established default, or add focused regression coverage
+as a courtesy. Re-run the full exact-head review and tests and tell the
+contributor what changed. Large rebases, behavior changes, architectural
+judgment, or newly discovered substantive defects remain contributor work.
@@ -30,6 +30,7 @@ npm run test:no-self-fitting
 npm run test:no-phantom-tools
 npm run test:gate
 npm run prompt:audit
+npm run pack:dry-run        # inspect the allowlisted package manifest; no workspace/private files
 ```

 ## 2. Local API smoke (loopback only)
@@ -67,8 +68,22 @@ degrade gracefully — they never fail the core run.

 ## 6. Publish

+- Confirm the release commit is already on `main`, CI is green at that exact
+  SHA, the version and changelog agree, and the tag is signed and points to that
+  commit. Never move or reuse a failed release tag; correct the release and use
+  a new version/tag.
 - Verify `repository.url` in `package.json` points at this repo. To retarget it:
  ```bash
  npm pkg set repository.url="git+https://github.com/<owner>/<repo>.git"
  ```
 - Tag the release only after Sections 1–2 are green.
+- Push the `v*` tag and wait for the tag workflow. It reruns
+  `npm run test:release`, the high-severity dependency audit, and the package
+  dry run against the exact tag, then retains the tested `.tgz`, its manifest,
+  and `SHA256SUMS` as workflow artifacts.
+- Inspect `pack-result.json`: only the allowlisted runtime, scripts, tools,
+  documentation, examples, and package metadata may ship. Workspace notes,
+  tests, AIWG/provider deployment internals, secrets, and local artifacts are
+  release blockers.
+- Publish the retained, checksum-matched package artifact. Do not rebuild a
+  different archive from another checkout after certification.
@@ -3442,9 +3442,9 @@
      "license": "MIT"
    },
    "node_modules/fast-uri": {
-      "version": "3.1.4",
-      "resolved": "https://registry.npmjs.org/fast-uri/-/fast-uri-3.1.4.tgz",
-      "integrity": "sha512-8JnbkQ4juDyvYs4mgFGQqg4yCYtFDtUtmp2QIQq11ZZe5CFQ5wcqm1rqDgAh/QdMySuBnPzMUiJUNZG5N/AiQw==",
+      "version": "3.1.6",
+      "resolved": "https://registry.npmjs.org/fast-uri/-/fast-uri-3.1.6.tgz",
+      "integrity": "sha512-7Ical1vFEMr0onbVzEDIreM22I4khW+fzyQPwvAFWBp1iwdshSZRsL4jjRvPG9JP1uiqMHRto+YU6R2/CzDz5Q==",
      "funding": [
        {
          "type": "github",
@@ -3801,9 +3801,9 @@
      }
    },
    "node_modules/hono": {
-      "version": "4.12.27",
-      "resolved": "https://registry.npmjs.org/hono/-/hono-4.12.27.tgz",
-      "integrity": "sha512-1yrb/+w6HWQJrUCLkJ2IF5jNIPvvFkblV5RNOYl6bV+OA6p9GLcMpHFFGTosSvHvcAUibuUukRqhlYI4z32C7Q==",
+      "version": "4.13.3",
+      "resolved": "https://registry.npmjs.org/hono/-/hono-4.13.3.tgz",
+      "integrity": "sha512-r8AO2mYHoLxSHkgafNeC/BXyb2vWRxD3jem4Ts+ptav8oTG5FIRifAjuJEmZI4bSvvc2ns0GxmIYiZnHqN3mMw==",
      "license": "MIT",
      "engines": {
        "node": ">=16.9.0"
@@ -4119,9 +4119,9 @@
      }
    },
    "node_modules/ip-address": {
-      "version": "10.2.0",
-      "resolved": "https://registry.npmjs.org/ip-address/-/ip-address-10.2.0.tgz",
-      "integrity": "sha512-/+S6j4E9AHvW9SWMSEY9Xfy66O5PWvVEJ08O0y5JGyEKQpojb0K0GKpz/v5HJ/G0vi3D2sjGK78119oXZeE0qA==",
+      "version": "10.5.0",
+      "resolved": "https://registry.npmjs.org/ip-address/-/ip-address-10.5.0.tgz",
+      "integrity": "sha512-R5SnVLJmgYYvf2F2ZgwSBnelz5G4q5AxIC277GDfUaNbrZKNANcBC7RHqYYePlszf4kBolVkJauG0ZjHHFh55g==",
      "license": "MIT",
      "engines": {
        "node": ">= 12"
@@ -4260,9 +4260,9 @@
      "license": "MIT"
    },
    "node_modules/js-yaml": {
-      "version": "4.3.0",
-      "resolved": "https://registry.npmjs.org/js-yaml/-/js-yaml-4.3.0.tgz",
-      "integrity": "sha512-1td788aAnnZ5qs7V2QIRl1owjtYpbKt749Y3xauqQgwIIGF/xXWz1wMTEBx5O3LK3lXLVuqXPdPxj2BoFHaW9Q==",
+      "version": "4.3.1",
+      "resolved": "https://registry.npmjs.org/js-yaml/-/js-yaml-4.3.1.tgz",
+      "integrity": "sha512-CY6crGq313MX8GkwvB7tzgp99vjQxY1++5y10/BKN/GUfHqWaOGQMNZkBvqSzsZKWk/ijwHlWzzkLulsGHhjWQ==",
      "dev": true,
      "funding": [
        {
@@ -4804,9 +4804,9 @@
      }
    },
    "node_modules/nanoid": {
-      "version": "3.3.16",
-      "resolved": "https://registry.npmjs.org/nanoid/-/nanoid-3.3.16.tgz",
-      "integrity": "sha512-bzlKTyNJ7+LdGIIwy8ijFpIqEQIvafahV7eYykJ8Cvh42EdJeODoJ6gUJXpQJvej1BddH8OqTXZNE/KfbWAu8Q==",
+      "version": "3.3.18",
+      "resolved": "https://registry.npmjs.org/nanoid/-/nanoid-3.3.18.tgz",
+      "integrity": "sha512-DTg4MJbGMWkfi6VZFdNt2/caMbQy4Ou+Op/hJQvGEWcnVfoA1QA+xzRKAzw9jD6+GVOOeYr/mIcuDSdug6F6+w==",
      "dev": true,
      "funding": [
        {
@@ -6018,9 +6018,9 @@
      }
    },
    "node_modules/undici": {
-      "version": "8.8.0",
-      "resolved": "https://registry.npmjs.org/undici/-/undici-8.8.0.tgz",
-      "integrity": "sha512-ubshXMXwF3MQIMF1y/WxZdNBnjEKeSg2wF5mcGUtU55YTw34tnVVpKRlLf7ruDXZ5344KokPVX4RBx1wJm64Bw==",
+      "version": "8.10.0",
+      "resolved": "https://registry.npmjs.org/undici/-/undici-8.10.0.tgz",
+      "integrity": "sha512-HvltHd7avK13QIw/oLe4qoOLyoVSoafqJ2jYOrtMRBkbYT31eiBQ8O0ehRKZiEZCMEyLFQNIADpgCWC5fALvYQ==",
      "license": "MIT",
      "engines": {
        "node": ">=22.19.0"
@@ -5,6 +5,17 @@
  "main": "dist/index.js",
  "types": "dist/index.d.ts",
  "type": "module",
+  "files": [
+    "dist/",
+    "scripts/",
+    "tools/",
+    "docs/",
+    "examples/",
+    ".env.example",
+    "docker-compose.yml",
+    "index.ts",
+    "tsconfig.json"
+  ],
  "scripts": {
    "build": "tsc",
    "start": "node dist/cli.js",
@@ -64,6 +75,8 @@
    "mcp": "tsx src/mcp-server.ts",
    "mcp:prod": "node dist/mcp-server.js",
    "test": "vitest run src && node scripts/test-ops-preflight.mjs && node scripts/test-model-matrix.mjs && node scripts/refusal-frontier.mjs --self-test",
+    "test:release": "npm run lint && npm run typecheck && npm test && npm run test:coverage && npm run doctor && npm run verify-claims && npm run test:no-fitting && npm run test:no-self-fitting && npm run test:gate && npm run prompt:audit && npm run smoke && npm run build",
+    "pack:dry-run": "npm pack --dry-run",
    "test:coverage": "vitest run --coverage",
    "test:watch": "vitest src",
    "lint": "eslint src/**/*.ts",
@@ -193,8 +193,9 @@ describe('detectLocalAgents — wiring: a well-known-dir CLI is reported install
    expect(claude?.installed).toBe(true);
    expect(claude?.path).toBe(claudePath);
    expect(claude?.version).toBe('1.2.3');       // parsed from the resolved binary's --version output
-    // codex/hermes are absent from this temp home → still not installed (no false positives).
-    expect(agents.find((a) => a.id === 'codex')?.installed).toBe(false);
+    // Do not assert that another real agent is absent: detection intentionally
+    // also scans system-wide install locations such as /usr/local/bin, so that
+    // result depends on the maintainer/CI host rather than this fixture.
  });

  it('detects authenticated OpenCode and Oh My Pi installations (#136)', async () => {