agentleFS
albertomartureloalbertomarturelo/sii/.agents/skills/cfd-upgrade/SKILL.md

cfd-upgrade

Skill19 starsChanged 7 days agoMITView on GitHub

Upgrade this project's CFD artifacts: read the installed version from the skills and the scaffold, fetch the CFD changelog, show what changed since, and apply each release's Upgrade steps in order after the user approves. Invoke explicitly; it changes committed files.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    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:

    If both versions match, say so and stop.

    line 34
  2. Distinctive:

    Apply, oldest entry first: Run each Upgrade step as written.

    line 44
  3. Distinctive:

    Do not skip an intermediate entry: a later release assumes the earlier Upgrade steps ran.

    line 59

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Upgrade the CFD skills and scaffold of this repository to the latest release, one changelog entry at a time. Nothing is changed before the user approves the plan.

  1. Installed version. Run:

    grep -h '^  cfd-version:' .agents/skills/*/SKILL.md | sort | uniq -c
    grep -h 'CFD version:' CLAUDE.md .claude/hooks/*.sh .github/workflows/*.yml 2>/dev/null | sort | uniq -c

    One version everywhere is the installed version. Several versions mean a partial upgrade: the installed version is the lowest. No stamp at all means a release older than 0.3.0: treat it as 0.2.0 if .agents/skills/ exists, 0.1.0 if procedures live in .claude/commands/.

  2. Latest version and changelog. Fetch:

    curl -fsSL https://raw.githubusercontent.com/albertomarturelo/context-first-development/main/.claude-plugin/plugin.json | grep '"version"'
    curl -fsSL https://raw.githubusercontent.com/albertomarturelo/context-first-development/main/CHANGELOG.md

    If both versions match, say so and stop. Otherwise keep only the entries newer than the installed version.

  3. Show the plan, then wait. For each pending entry, oldest first: its version, its Added/Changed/Removed lines, and its Upgrade steps verbatim. Flag every step that touches a file this project has edited since installing it (git log --oneline -- <path> shows commits other than the install). Ask for an explicit yes before step 4.
  4. Apply, oldest entry first. Run each Upgrade step as written. Three rules override any step:
    • CLAUDE.md, AGENTS.md and docs/ are the project's own context: apply the described edit by hand, never overwrite them.
    • A hook, workflow or skill the project edited: show the diff against the new template and let the user choose per file.
    • Skills installed through the Claude Code plugin (/cfd:*) or npx skills add are updated by their installer, as the entry says; apply only the scaffold steps here.
  5. Verify and commit. The checks from step 1 must print one version, the latest. Then run /validate-context. Commit the upgrade on its own, with the versions crossed and the entries applied in the message, and no code changes mixed in.

Do not skip an intermediate entry: a later release assumes the earlier Upgrade steps ran. Do not downgrade. If the changelog cannot be fetched, stop and say so; do not guess the steps.

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.