agentleFS
Sign inStart free
Wsh7AshWsh7Ash/skillmcp/skills/replay-transcript-for-regression/SKILL.md

replay-transcript-for-regression

Skill0 starsChanged 14 days agoMITView on GitHub

Re-run a stored session against a new build to catch protocol drift before a client does. Use when the server is about to be released and its wire behaviour must not change

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:

    Run it against the previous build and store the output as the baseline

    line 21
  2. Distinctive:

    Run it against the new build with identical input

    line 22
  3. Distinctive:

    every difference is either explained in the changelog or fixed

    line 63

≈ “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

Replay Transcript For Regression

Re-run a stored session against a new build to catch protocol drift before a client does

Use when: the server is about to be released and its wire behaviour must not change

Avoid when: the release intentionally changes the protocol and new fixtures were written

Procedure

  1. Pick the transcript that covers the most client-visible behaviour
  2. Run it against the previous build and store the output as the baseline
  3. Run it against the new build with identical input
  4. Compare semantically, ignoring declared volatile fields
  5. Investigate every difference before deciding it is intended
  6. Publish the new baseline only with the release notes that explain it

Signals

  • the replay completes without a timeout or crash
  • differences are limited to fields the release intended to change
  • the baseline file is committed with the release
  • a deliberate change appears in the changelog

Patterns

1. Baseline pair

transcript-before.jsonl and transcript-after.jsonl

2. Semantic compare

parse both, drop volatile fields, diff

3. Explained diff

new field 'instructions' added in 1.1.0

Pitfalls

  • Comparing raw bytes and failing on key ordering
  • Accepting a diff without checking it against the release intent
  • Replaying without a timeout so a hung build blocks the release

Done when

  • a baseline exists for the previous build and is compared semantically
  • every difference is either explained in the changelog or fixed

References

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.

Check your setup, free