Update AI Team Orchestration with risk-based delivery v2 🤖🤖🤖

[denis-a-evdokimov]

Pull Request Checklist

• I have read and followed the CONTRIBUTING.md guidelines.
• I have read and followed the Guidance for submissions involving paid services. (No paid service is introduced.)
• My contribution adds or updates agent, skill, and plugin files in the correct directories.
• The files follow the required naming conventions.
• The content is clearly structured and follows the existing plugin format.
• The workflow content was exercised and independently reviewed with GitHub Copilot.
• I ran npm start and verified the generated marketplace/docs outputs are up to date.
• I am targeting the main branch for this pull request.

---

Description

Updates the existing AI Team Orchestration plugin to the canonical v2 risk-based delivery workflow:

• freezes the application candidate at push and binds all selected evidence to its full commit ID
• lets projects select independent review, QA, and post-merge checks in proportion to risk while requiring concrete checks for every code/configuration change
• routes blocked findings through a Producer-owned Delivery Ledger and scoped Branch Reopen Packet
• makes merge authorization atomic with the Candidate ID (expected-head merge or equivalent protected queue revalidation)
• adds explicit capability-vs-authority and untrusted-content boundaries for inherited built-in, MCP, and extension tools
• adds Safe Git value grammar, verified endpoints, fixed command forms, destination confirmation, and one remote-add action per distinct missing mapping
• keeps the stable ai-team-dev, ai-team-producer, and ai-team-qa IDs while intentionally omitting tools and model
• adds canonical delivery-workflow and safe-git reference files
• updates the Awesome-owned marketplace README and generated marketplace/docs indexes

The managed content was exported one way from the merged canonical repository using its verified non-mutating patch-preparation flow.

---

Type of Contribution

• New instruction file.
• New prompt file.
• New agent file.
• New plugin.
• New skill reference files.
• New agentic workflow.
• New canvas extension.
• Update to existing agent, plugin, and skill content.
• Other (please specify):

---

Validation

• Canonical source PRs: feat: harden risk-based AI team delivery denis-a-evdokimov/ai-team-orchestration#7 and fix: make merge and remote setup race-safe denis-a-evdokimov/ai-team-orchestration#8
• Canonical final merge commit: 0108a72a27fac358b3b3e65442f3beceecd35d3b
• Awesome base: 30472ecf0fe34cc561df958c08501ecc5ca80ea4
• Awesome candidate: f149776ce1e561b03c69b6ca223f6f6d4c8fe256
• Canonical npm run validate: PASS
• Canonical npm test: PASS, 99/99
• Canonical serial tests: PASS, 99/99
• npm run skill:validate: PASS
• npm run plugin:validate: PASS
• npm start: PASS
• git diff --check: PASS
• Canonical-to-Awesome alignment check after commit: PASS
• Independent canonical and downstream reviews: PASS, no blocker/major finding
• Both Copilot review threads addressed, replied to, and resolved

---

Additional Notes

This supersedes closed PR #2275 with a fresh branch based directly on current upstream/main. The plugin version remains 2.0.0 because v2 was not published by the closed PR. Independent review, QA, and smoke checks are intentionally risk-selected rather than universally mandatory; high-risk changes still require applicable security evidence or explicit maintainer risk acceptance.

By submitting this pull request, I confirm that my contribution abides by the Code of Conduct and will be licensed under the MIT License.

[github-actions]

🔒 PR Risk Scan Results

Scanned 16 changed file(s).

Severity	Count
🔴 High	0
🟠 Medium	1
ℹ️ Info	0

Severity	Rule	File	Line	Match
🟠	package-exec-command	docs/README.skills.md	31	| [acreadiness-assess](../skills/acreadiness-assess/SKILL.md)<br />`gh skills install github/awesome-copilot acreadiness-assess` | Run the AgentRC readiness assessment on the curre

This is an automated soft-gate report. Findings indicate review targets and do not block merge by themselves.

[github-actions]

🔍 Vally Lint Results

⚠️ Warnings or advisories found

Scope	Checked
Skills	1
Agents	3
Total	4

Severity	Count
❌ Errors	0
⚠️ Warnings	0
ℹ️ Advisories	1

Summary

Level	Finding
ℹ️	Vally currently lints SKILL.md content. Agent files were detected but skipped:

Full linter output

### Linting skills/ai-team-orchestration
npm warn EBADENGINE Unsupported engine {
npm warn EBADENGINE   package: 'commander@15.0.0',
npm warn EBADENGINE   required: { node: '>=22.12.0' },
npm warn EBADENGINE   current: { node: 'v20.20.2', npm: '10.8.2' }
npm warn EBADENGINE }
npm warn deprecated prebuild-install@7.1.3: No longer maintained. Please contact the author of the relevant native addon; alternatives are available.
✅ ai-team-orchestration (2/2 checks passed)
    ✓ [spec-compliance] All 1 skill(s) are spec-compliant.
        ✓ spec-compliance: All spec checks passed.
    ✓ [valid-refs] All file references across 1 skill(s) are valid.
        ✓ valid-refs: All file references resolve to existing files within the skill directory.

1 skill(s) linted, 1 passed

### Agent files detected (not linted by vally)
ℹ️ Vally currently lints SKILL.md content. Agent files were detected but skipped:
agents/ai-team-dev.agent.md
agents/ai-team-producer.agent.md
agents/ai-team-qa.agent.md

[denis-a-evdokimov]

Automated risk-scan triage: the single medium package-exec-command match is an unchanged, pre-existing gh skills install entry for acreadiness-assess in generated docs/README.skills.md. This PR's generated hunk only updates the ai-team-orchestration row and adds its two reference links. No source or generated-file change is appropriate; all high-severity findings are zero and the soft gate is non-blocking.

[Copilot]

Pull request overview

Updates AI Team Orchestration to a risk-based v2 delivery workflow.

Changes:

• Adds frozen-candidate, evidence-binding, and scoped-reopen processes.
• Introduces canonical delivery and Safe Git references.
• Updates agents, plugin metadata, and generated indexes.

Reviewed changes

Copilot reviewed 16 out of 16 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
agents/ai-team-dev.agent.md	Defines Dev lifecycle and boundaries.
agents/ai-team-producer.agent.md	Adds risk-based gate orchestration.
agents/ai-team-qa.agent.md	Adds optional candidate-bound QA.
skills/ai-team-orchestration/SKILL.md	Updates the skill workflow.
skills/ai-team-orchestration/references/anti-patterns.md	Documents delivery anti-patterns.
skills/ai-team-orchestration/references/brainstorm-format.md	Clarifies simulated perspectives.
skills/ai-team-orchestration/references/delivery-workflow.md	Adds the canonical delivery lifecycle.
skills/ai-team-orchestration/references/project-brief-template.md	Adds delivery governance sections.
skills/ai-team-orchestration/references/safe-git-values.md	Adds validated Git procedures.
skills/ai-team-orchestration/references/sprint-plan-template.md	Adds risk, gate, and evidence fields.
plugins/ai-team-orchestration/README.md	Documents v2 usage.
plugins/ai-team-orchestration/.github/plugin/plugin.json	Updates metadata to v2.0.0.
.github/plugin/marketplace.json	Regenerates marketplace metadata.
docs/README.agents.md	Updates generated agent entries.
docs/README.plugins.md	Updates the plugin index.
docs/README.skills.md	Updates the skill and reference index.

[Copilot]

Pull request overview

Copilot reviewed 16 out of 16 changed files in this pull request and generated 7 comments.

[aaronpowell]

There are some high and medium level review comments from Copilot, do you want to review them before merging

[denis-a-evdokimov]

There are some high and medium level review comments from Copilot, do you want to review them before merging

I think it is fine to merge.
I will check over the weekend if any follow up changes are required.

[Copilot]

Pull request overview

Copilot reviewed 16 out of 16 changed files in this pull request and generated 1 comment.

[Copilot]

Pull request overview

Copilot reviewed 16 out of 16 changed files in this pull request and generated 3 comments.

[Copilot]

Pull request overview

Copilot reviewed 16 out of 16 changed files in this pull request and generated no new comments.

Comments suppressed due to low confidence (1)

skills/ai-team-orchestration/references/delivery-workflow.md:123

• The expected-head guard only protects the PR head; it does not protect the target branch revision. If the target branch advances after the candidate checks, GitHub can still create a regular merge against that newer base (when conflict-free), so the resulting merge commit has never been covered by the candidate-bound evidence. Record the base commit in the ledger and invalidate Ready when it moves, or require strict up-to-date/merge-queue checks that rerun against the latest-base merge result before allowing the merge.

Use a regular merge, but make the merge operation itself atomically require the application head to equal the Delivery Ledger Candidate ID. For GitHub, use an expected-head merge such as `--match-head-commit <Candidate ID>` or the merge API's `sha` field. A protected merge queue is acceptable only when it revalidates candidate-bound checks and rejects or requeues unexpected head movement. A separate comparison immediately before an unguarded merge is insufficient. If the guard fails or the queue observes another head, move to Hold and reconcile the new candidate before any merge. Run selected post-merge checks against the merged artifact.

[Copilot]

Pull request overview

Copilot reviewed 16 out of 16 changed files in this pull request and generated 3 comments.

Comments suppressed due to low confidence (1)

skills/ai-team-orchestration/references/safe-git-values.md:75

• The baseline can accept a clean but incomplete worktree. This override suppresses required filters such as Git LFS (leaving pointer files), while --no-recurse-submodules leaves gitlinks uninitialized; ls-files and status still pass, and no check-attr/gitlink detection stops the setup as the paragraph promises. Detect these unsupported repository requirements without executing them and hand off for reviewed setup before declaring the clone verified.

The clone deliberately leaves the worktree empty. Its immediate `core.fsmonitor` check is another expected non-match and catches destination-conditional includes that were inactive in the parent directory; do not checkout if it returns a value. Before the checkout command, use a trusted filesystem API—not shell redirection or repository-provided commands—to create `CLONE_DESTINATION/.git/info` if absent and then create its regular `attributes` file with exactly this one LF-terminated line: `* -text -filter -diff -ident -working-tree-encoding`. Verify the directory and file are not links and the file reads back exactly. This highest-precedence private attribute rule prevents repository `.gitattributes` from activating configured clean/smudge/process filters, diff drivers, ident expansion, or working-tree encodings during checkout and later switches.