agentleFS
ArthurZakirovArthurZakirov/SystemSmith/skills/change-consistency/SKILL.md

change-consistency

Skill0 starsChanged 6 months agoMITView on GitHub

Use when implementing or revising behavior, terminology, interfaces, or usage patterns, or reviewing a branch for changes that were not carried through consistently.

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:

    Decide from purpose and evidence: Ask whether each affected element still serves a supported purpose and accurately expresses the intended …

    line 29
  2. Distinctive:

    Update inaccurate representations, including test names after assertion changes.

    line 31
  3. Distinctive:

    Verify behavior and completeness separately: Run applicable checks, then review the complete branch diff and affected relationships for …

    line 44

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

🔄 Complete the consequences of a change

Complete the consequences of the accepted change across affected code and supporting artifacts, including places outside the initially edited files. Preserve unrelated pre-existing issues. Apply this procedure again to follow-up changes, regardless of where the requirement originated.

  1. Establish before and after. Identify the branch's starting point: the most recent commit shared with the branch it will merge into, unless a comparison commit was supplied. Compare it with the current state, including uncommitted work. State what behavior, meaning, or contract changes and identify the intended-use entry points. These may be application entry points, end-to-end scenarios, or library consumers. Distinguish consequences of this branch from pre-existing issues, even within edited files.
  2. Follow relationships in both directions. Inspect consumers needing adaptation and dependencies that may become unnecessary, including functions inside still-used modules. Follow representations of the same concept: names, assertions, fixtures, configuration, documentation, and examples. These are search directions, not an exhaustive checklist. Trace assumptions and equivalent meanings even when their names differ; inspect occurrences rather than relying on text matches or reference counts.
  3. Decide from purpose and evidence. Ask whether each affected element still serves a supported purpose and accurately expresses the intended behavior. Update inaccurate representations, including test names after assertion changes. Remove newly unnecessary implementation together with obsolete tests and dependencies. Unit-test callers alone do not establish a continuing intended use. Check supported external interfaces, framework discovery, test infrastructure, intentional historical references, and required compatibility before changing or removing them.
  4. Repeat after consequential edits. Removing one element can make its callers, tests, imports, fixtures, or configuration obsolete. Continue until another pass identifies no unresolved consequences. Report uncertainty about dynamic or external usage rather than inventing a reason to retain or delete something. Keep the review within the change's causal scope.
  5. Verify behavior and completeness separately. Run applicable checks, then review the complete branch diff and affected relationships for superseded assumptions. Passing tests, remaining references, or a clean search alone do not establish completeness. Briefly report consequential updates beyond the initial edit and any remaining uncertainty; limit coverage claims to what was actually inspected.

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.