agentleFS
gnegreiros-uxgnegreiros-ux/agentica-design-system/.agents/skills/post-change-pipeline/SKILL.md

post-change-pipeline

Skill2 starsChanged 46 days agoApache-2.0View on GitHub

Mandatory impact analysis before any commit to the Agentica repository — identifies which files changed, what downstream updates that requires (site rebuild, ADR, CSS token regeneration), presents the impact as a report, and waits for explicit human approval before executing anything. Use before every commit, however small.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run git diff --name-only and git diff --cached --name-only to identify modified files.

    line 21
  2. Distinctive:

    Create the approved ADRs

    line 70

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Skill: post-change-pipeline

Cross-tool port of .claude/skills/post-change-pipeline/SKILL.md (Claude Code's native implementation) — see governance/ai-skills-reference.md for the full picture of what Agentica's skills automate. This file follows the open Agent Skills format (agentskills.io) so Codex CLI and GitHub Copilot CLI load it automatically from .agents/skills/; content is identical in substance to the Claude Code version.

Trigger

This pipeline runs mandatorily after every modification and before every commit. It cannot be skipped, even for a minor change.

Step 1 — Impact analysis (automatic)

Run git diff --name-only and git diff --cached --name-only to identify modified files.

Apply the impact matrix:

Modified filesUpdates to evaluate
tokens/primitives.jsonSite rebuild, GitHub Projects entry, ADR if new token or palette
tokens/semantic.jsonSite rebuild, GitHub Projects entry, generated tokens.css
tokens/component.jsonSite rebuild, GitHub Projects entry, Principal Designer approval required
site/build.jsSite rebuild only
site/dist/GitHub Projects entry only (dist is an output)
guidelines/ or components/Site rebuild, GitHub Projects entry
decisions/ADR-*.mdSite rebuild (ADR page), GitHub Projects entry
governance/rules/ or .claude/skills/ / .agents/skills/GitHub Projects entry
AGENTS.md, DESIGN.md, README.mdGitHub Projects entry

Task tracking (status, domain) lives exclusively in GitHub Projects (ADR-069) — not in a versioned file in the repo.

Step 2 — Impact report (present to the human)

Present a structured report before any commit:

## Change impact — approval required

### Modified files
- [list of files from git diff]

### Proposed updates
- [ ] Site rebuild (`node site/build.js`) — [reason]
- [ ] ADR-XXX to create — [proposed title] — if architectural decision
- [ ] CSS tokens to regenerate — if tokens modified
- [ ] Other: [description]

### Points of attention
- [Any component token modification → approval reminder]
- [Any new token → suggest an ADR]

Do not commit before an explicit response from the human.

Step 3 — Execution (only after approval)

Execute only the approved tasks, in this order:

  1. Regenerate CSS tokens if tokens/ was modified
  2. Rebuild the site if approved (node site/build.js)
  3. Create the approved ADRs
  4. Stage and commit all files in a single coherent commit

Escalation rules

  • Modification of tokens/component.json → explicitly mention that Principal Designer approval is required
  • New primitive or semantic token → propose an ADR
  • Removal of a token → block and request a full impact audit

Anti-patterns

❌ Committing without presenting the impact report
❌ Rebuilding the site without it being approved
❌ Updating GitHub Projects after the commit rather than before
❌ Proposing all updates even when not relevant (use the matrix)
❌ Waiting for the human to ask for the GitHub Projects update — always propose it

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.