agentsclimarketplace

Api changelog impact analyzer

Skill sisodiabhumca/agent-skills/skills/api-changelog-impact-analyzer

Vendor-neutral skill for analyzing an API changelog and identifying likely breaking changes and downstream impacts.From its SKILL.md

Install
npx -y skills add sisodiabhumca/agent-skills --skill api-changelog-impact-analyzer

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

1.3 KB, 258 tokens by cl100k_base, as published. Nobody here has run it

When to invoke

  • You have API release notes/changelog text (Markdown) from any provider.
  • You need to identify potentially breaking changes and what client code to review.
  • You want a vendor-neutral heuristic analysis (not tied to any specific API platform).

Inputs needed

  • Path to a changelog markdown/text file.
  • (Optional) A list of client endpoints your app uses (JSON) to cross-check.

Workflow

  1. Parse changelog into bullet items and headings.
  2. Classify each item into:
    • breaking (removal, rename, required fields, auth changes)
    • behavior change (defaults, limits, ordering)
    • additive (new fields/endpoints)
    • deprecation
  3. Extract impacted surfaces (endpoint paths, field names) via regex heuristics.
  4. If a client-usage file is provided, highlight overlaps.
  5. Output a prioritized action list.

Output format

  • Markdown summary to stdout.
  • JSON report (optional) with categorized items and detected impacts.

Guardrails

  • Heuristic only; do not claim completeness.
  • Do not infer security posture; only flag explicit auth/scope changes mentioned.

Reference code

  • api_changelog_impact_analyzer.py

What ships with it: 2 files

7.6 KB alongside SKILL.md, 1 of them executable

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.