00PrabalK00/api-regression-guard
skills/api-regression-guard/SKILL.md
Check whether code changes silently broke API contracts, behavior, schemas, or compatibility expectations. Use when reviewing endpoint changes, SDK updates, or public interface diffs.
development
Updated Mar 23, 2026
$ install --global
skillsauthnpx skillsauth add 00PrabalK00/claude-skills api-regression-guardInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
4 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
4
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: Mar 23, 2026, 11:31 PM140.2s2 files scanned
SKILL.md
- name:
- api-regression-guard
- description:
- Check whether code changes silently broke API contracts, behavior, schemas, or compatibility expectations. Use when reviewing endpoint changes, SDK updates, or public interface diffs.
API Regression Guard
Overview
Check whether code changes silently broke API contracts, behavior, schemas, or compatibility expectations.
Core Workflow
- Inspect the relevant diffs, configs, source files, or artifacts before forming a conclusion.
- Compare the current state against repo conventions, expected guarantees, and known risk areas.
- Prioritize the highest-impact gaps, regressions, or improvement opportunities instead of surface-level noise.
- Recommend the smallest high-leverage changes and the checks that would validate them.
Deliver
- Ranked findings or improvement opportunities with clear evidence.
- The main risks, tradeoffs, or regressions that matter most.
- Validation steps or follow-up edits that would reduce risk.
Guardrails
- Lead with findings, not generic praise or low-value commentary.
- Use repo evidence and surrounding context instead of reviewing in isolation.
- Flag missing validation when a recommendation depends on behavior you could not confirm.
- Prefer repository evidence, command output, and nearby tests over vague intuition.
Related Skills
00PrabalK00/vendor-evaluation-scorecard
tools
VerifiedTrustedCommunity
Compare vendors or tools using weighted criteria, tradeoffs, risks, and recommendation logic. Use when selecting platforms or partners.
SKILL.mdUpdated Mar 23, 2026
00PrabalK00/user-interview-synthesizer
data-ai
VerifiedTrustedCommunity
Extract needs, pain points, quotes, and behavioral patterns from user interviews. Use when synthesizing qualitative research.
SKILL.mdUpdated Mar 23, 2026
00PrabalK00/user-feedback-clusterer
databases
VerifiedTrustedCommunity
Group user feedback into themes, bugs, requests, sentiment, and confusion points. Use when analyzing feedback corpora.
SKILL.mdUpdated Mar 23, 2026
00PrabalK00/test-writer-coverage-closer
testing
VerifiedTrustedCommunity
Find missing coverage, design useful unit or integration tests, add fixtures or mocks, and explain remaining edge cases. Use when a feature lacks tests, a bug needs regression coverage, or coverage gaps block confidence.
SKILL.mdUpdated Mar 23, 2026