Skip to content

🤖🤖🤖 Update quality-playbook skill to v1.5.6 + add agent#1402

Merged
aaronpowell merged 1 commit into
github:stagedfrom
andrewstellman:quality-playbook-v1.4.0
May 11, 2026
Merged

🤖🤖🤖 Update quality-playbook skill to v1.5.6 + add agent#1402
aaronpowell merged 1 commit into
github:stagedfrom
andrewstellman:quality-playbook-v1.4.0

Conversation

@andrewstellman
Copy link
Copy Markdown
Contributor

@andrewstellman andrewstellman commented Apr 15, 2026 •
edited
Loading

Summary

  • Updates quality-playbook skill from v1.3.x to v1.5.6

Changes from 1.3.x to 1.4.2:

  • Adds agents/quality-playbook.agent.md — checks skill installation, explains phases, handles "keep going" / "help" / "what happened"
  • Six-phase interactive architecture with end-of-phase messages
  • Four iteration strategies (gap, unfiltered, parity, adversarial)
  • TDD red-green verification for every confirmed bug
  • Quality gate script for mechanical artifact validation
  • Moved ITERATION.md into references/ (installed via wildcard copy)

Updated to v1.5.6.
Changes since v1.4.2:

SKILL.md modernized (skills bundle now includes phase_prompts/, agents/, and bin/citation_verifier.py per the v1.5.6 install contract; quality_gate.sh retired in v1.4.5 and replaced by quality_gate.py).
references/ expanded from 12 to 16 files.
The top-level agents/quality-playbook.agent.md now installs from awesome-copilot's own skills/quality-playbook/ folder instead of pointing users to the upstream Quality Playbook repo, so adopters discovering the skill via awesome-copilot can install without leaving the marketplace. Manual cp recipes provided for all four supported AI tools (Copilot, Cursor, Claude Code, Continue). Pointer to the upstream repo preserved as alternative for users who want the full script-driven install UX.

Empirically validated against Cursor, Claude Code, and GitHub Copilot install flows on 2026-05-08.

See: https://www.oreilly.com/radar/ai-is-writing-our-code-faster-than-we-can-verify-it/

Copilot AI review requested due to automatic review settings April 15, 2026 14:34
@github-actions
Copy link
Copy Markdown
Contributor

github-actions Bot commented Apr 15, 2026 •
edited
Loading

🔍 Skill Validator Results

⚠️ Warnings or advisories found

Scope Checked
Skills 1
Agents 1
Total 2
Severity Count
--- ---:
❌ Errors 0
⚠️ Warnings 1
ℹ️ Advisories 0

Summary

Level Finding
ℹ️ Found 1 skill(s)
ℹ️ [quality-playbook] 📊 quality-playbook: 66,332 BPE tokens [chars/4: 73,092] (comprehensive ✗), 128 sections, 44 code blocks
ℹ️ [quality-playbook] ⚠ Skill is 66,332 BPE tokens (chars/4 estimate: 73,092) — "comprehensive" skills hurt performance by 2.9pp on average. Consider splitting into 2–3 focused skills.
ℹ️ Found 1 agent(s)
ℹ️ Validated 1 agent(s)
ℹ️ ✅ All checks passed (1 skill(s), 1 agent(s))
Full validator output ```text Found 1 skill(s) [quality-playbook] 📊 quality-playbook: 66,332 BPE tokens [chars/4: 73,092] (comprehensive ✗), 128 sections, 44 code blocks [quality-playbook] ⚠ Skill is 66,332 BPE tokens (chars/4 estimate: 73,092) — "comprehensive" skills hurt performance by 2.9pp on average. Consider splitting into 2–3 focused skills. Found 1 agent(s) Validated 1 agent(s)

✅ All checks passed (1 skill(s), 1 agent(s))

</details>

@andrewstellman andrewstellman force-pushed the quality-playbook-v1.4.0 branch from a2b9013 to 57eae8d Compare April 15, 2026 14:42
Copilot AI reviewed Apr 15, 2026
View reviewed changes
Copy link
Copy Markdown
Contributor

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Copilot encountered an error and was unable to review this pull request. You can try again by re-requesting a review.

aaronpowell
aaronpowell requested changes Apr 28, 2026
View reviewed changes
Comment thread skills/quality-playbook/quality_gate.sh Outdated
Comment thread skills/quality-playbook/quality_gate.sh Outdated
Comment thread agents/quality-playbook.agent.md Outdated
@andrewstellman
Copy link
Copy Markdown
Contributor Author

andrewstellman commented Apr 28, 2026 via email

@aaronpowell
Copy link
Copy Markdown
Contributor

aaronpowell commented Apr 29, 2026

fine to iterate on this PR. We squash merge so "messy" history isn't really a problem

andrewstellman added a commit to andrewstellman/awesome-copilot that referenced this pull request May 8, 2026
@andrewstellman
Update quality-playbook to v1.5.6
a79bd6a 
- SKILL.md updated to v1.5.6 (2738 lines, was 2322 at v1.4.2). Banner
  + version stamps + JSON schema examples all bumped.
- references/ expanded to 16 files (was 12). New: challenge_gate.md,
  code-only-mode.md, orchestrator_protocol.md, run_state_schema.md.
  Existing files updated to current QPB content.
- phase_prompts/ added (9 files). New v1.5.x directory carrying
  externalized phase prompts the runner uses.
- agents/ added inside the skill bundle (3 files: quality-playbook.agent.md,
  quality-playbook-claude.agent.md, calibration_orchestrator.md). The
  v1.5.6 install bundle includes these so README's Step 4
  'claude --agent agents/...' invocation resolves from the install
  destination.
- bin/citation_verifier.py added — required by quality_gate.py for
  byte-equality citation verification (per QPB BUG-005 closure).
- quality_gate.sh retired in v1.4.5; replaced with quality_gate.py
  (Python rewrite with json.load + 108-test unit suite at upstream).

agents/quality-playbook.agent.md (top-level awesome-copilot agent file)
install instructions rewritten to install from awesome-copilot's own
skills/quality-playbook/ folder rather than from the andrewstellman/
quality-playbook upstream — adopters discovering the skill via
awesome-copilot can install without leaving the marketplace. Locations
list expanded to all four supported tools (Copilot, Cursor, Claude
Code, Continue). Pointer to upstream preserved as alternative for
users who want the full script-driven install UX (auto-detect, marker
directory creation, smoke checks).

Closes github#1402.
@github-actions github-actions Bot added agent PR touches agents new-submission PR adds at least one new contribution skills PR touches skills labels May 8, 2026
@andrewstellman andrewstellman requested a review from dvelton as a code owner May 9, 2026 01:22
@github-actions github-actions Bot added branched-main PR appears to include plugin files materialized from main skill-check-error Skill validator reported errors skill-check-warning Skill validator reported warnings and removed skills PR touches skills agent PR touches agents new-submission PR adds at least one new contribution labels May 9, 2026
andrewstellman added a commit to andrewstellman/awesome-copilot that referenced this pull request May 10, 2026
@andrewstellman
fix: agent name field to lowercase-hyphenated form (validator complia…
bedbe84 
…nce)

PR github#1402 validator flagged 'name: "Quality Playbook"' as failing two rules:
  - Agent name must match filename (expected 'Quality Playbook.agent.md')
  - Agent name must be lowercase alphanumeric and hyphens only

Change name to 'quality-playbook' in both the top-level agent file and
the bundle copy under skills/quality-playbook/agents/. The bundle's
quality-playbook-claude.agent.md was already correct.
@andrewstellman andrewstellman changed the title 🤖🤖🤖 Update quality-playbook skill to v1.4.0 + add agent 🤖🤖🤖 Update quality-playbook skill to v1.5.6 + add agent May 10, 2026
@andrewstellman andrewstellman force-pushed the quality-playbook-v1.4.0 branch from bedbe84 to ce0a291 Compare May 10, 2026 17:39
@github-actions github-actions Bot added agent PR touches agents new-submission PR adds at least one new contribution skills PR touches skills and removed branched-main PR appears to include plugin files materialized from main skill-check-error Skill validator reported errors labels May 10, 2026
@andrewstellman
Update quality-playbook skill to v1.5.6 + add agent
9f06edd 
Rebuilds branch from upstream/staged (was previously merged from
upstream/main, which brought in materialized plugin files that
fail Check Plugin Structure on PRs targeting staged).

Changes vs. staged:
- Update skills/quality-playbook/ to v1.5.6 (31 bundled assets:
  SKILL.md + LICENSE.txt + 16 references/ + 9 phase_prompts/ +
  3 agents/ + bin/citation_verifier.py + quality_gate.py).
- Add agents/quality-playbook.agent.md (top-level orchestrator).
  name: quality-playbook (validator-compliant).
- Update docs/README.skills.md quality-playbook row description
  + bundled-assets list to v1.5.6.
- Fix 'unparseable' → 'unparsable' in quality_gate.py (5 instances;
  codespell preference, both spellings valid).

Closes the v1.4.0 → v1.5.6 update in a single clean commit on top of
upstream/staged. The preserved backup branch backup-bedbe84-pre-rebuild
(SHA bedbe84) holds the prior history for reference.
@andrewstellman andrewstellman force-pushed the quality-playbook-v1.4.0 branch from ce0a291 to 9f06edd Compare May 10, 2026 17:43
@andrewstellman
Copy link
Copy Markdown
Contributor Author

andrewstellman commented May 10, 2026

Force-pushed to rebuild this branch on top of upstream/staged instead of the previously-merged upstream/main. The earlier merge (commit 8553ea7) brought in materialized plugin files that fail Check Plugin Structure on staged-targeted PRs — those are auto-copied to main at publish time and shouldn't appear in the diff for staged.
The branch now contains a single clean commit applying the v1.4.x → v1.5.6 quality-playbook update on top of staged. No QPB-related changes were lost; the prior history is preserved locally in backup-bedbe84-pre-rebuild for reference.
Also fixed in this rebuild: unparseable → unparsable in 5 instances in skills/quality-playbook/quality_gate.py (codespell preference; both spellings are valid English).

@aaronpowell aaronpowell merged commit b8441d2 into github:staged May 11, 2026
12 checks passed
@aaronpowell
Copy link
Copy Markdown
Contributor

aaronpowell commented May 11, 2026

I'll merge this in as-is, but it's worth considering the validator feedback for a future iteration.

The skill itself is big, there's a lot of tokens that it'll (potentially) use in what it does, which could be detrimental to performance and behaviour.

@andrewstellman
Copy link
Copy Markdown
Contributor Author

andrewstellman commented May 11, 2026 via email

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

Copilot code review Copilot Copilot left review comments

@aaronpowell aaronpowell aaronpowell approved these changes

@dvelton dvelton Awaiting requested review from dvelton

Assignees

No one assigned

Labels

agent PR touches agents new-submission PR adds at least one new contribution skill-check-warning Skill validator reported warnings skills PR touches skills

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@andrewstellman @aaronpowell