verify

Classifies diagnosed code behavior as bug, intended behavior, or unclear using repository files and tests.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/ahliweb/sikesra --skill verify-ahliweb
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: verify
Source: https://github.com/ahliweb/sikesra/tree/main/awcmsmicro-dev/.flue/skills/verify
Command: npx skills add https://github.com/ahliweb/sikesra --skill verify-ahliweb

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps decide whether a reported code issue is truly a bug, an intended behavior, or still unclear, so fix work only starts when the evidence supports it.

Core Features & Use Cases

  • Bug Triage: Compare the diagnosed symptom against surrounding code, comments, and call sites to separate defects from deliberate behavior.
  • Documentation Cross-Check: Read repository docs, AGENTS.md, CONTRIBUTING.md, and package READMEs to confirm whether the behavior is documented and expected.
  • Test Evidence Review: Inspect existing tests to see whether they reinforce the current behavior or reveal that the test suite itself is wrong.
  • Use Case: A maintainer receives a report about unexpected API output and uses this Skill to determine whether the code should be fixed or the reporter should be pointed to the design docs.

Quick Start

Use the verify skill to assess the diagnosed issue against the surrounding code, documentation, and tests, then return a verdict of bug, intended-behavior, or unclear.

Frequently Asked Questions about verify

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify if a reported code issue is actually a bug or intended behavior?

To verify if a reported code issue is a bug, compare the diagnosed symptom against surrounding source files, documentation, and tests. This read-only review cross-checks code intent against repository docs to classify behavior as a defect, intended, or unclear before starting a fix.

What is the best way to triage unexpected API output against design docs?

The best way to triage unexpected API output is to perform a read-only review of the surrounding code, package READMEs, and design docs. This documentation cross-check confirms whether the behavior is documented and expected or requires a fix.

How does reviewing existing tests help determine if a test suite is wrong?

Reviewing existing tests helps determine if a suite is wrong by inspecting whether the tests reinforce the current code behavior or contradict it. This test evidence review reveals if the unexpected symptom is a defect or if the test itself needs updating.

Can I use bug triage analysis to modify source files and execute tests directly?

No, bug triage analysis requires a read-only review of source files, documentation, and tests. It avoids edits, execution, or external writes to ensure maintainers safely compare symptoms against code intent before allowing any fix work to begin.

When do I need documentation cross-checks for repository triage cases?

You need documentation cross-checks for repository triage cases when maintainers must compare reported symptoms against AGENTS.md, CONTRIBUTING.md, and package READMEs. This confirms whether the diagnosed code behavior is documented and expected before approving a fix.