| name | analyze-build |
| description | Full pipeline — gather failures, analyze, and search for known fixes |
| user-invocable | true |
Analyze Build
Full pipeline that gathers test failure data from a Jenkins build, analyzes failures with test and product code context, and searches for known fixes.
Usage
/analyze-build <job-path> --build <number> [--limit N] [--type TYPE]
<job-path> (required) — full Jenkins job path: <team>/<ipv4|ipv6>/<component> (e.g. proton/ipv4/sat-stream-rhel9-Insights-Advisor)
--build <number> (required) — the Jenkins build number (e.g. --build 35)
Examples:
/analyze-build proton/ipv4/sat-stream-rhel9-Insights-Advisor --build 35
/analyze-build proton/ipv4/sat-stream-rhel9-RHCloud --build 1019
/analyze-build proton/ipv4/sat-stream-rhel9-RHCloud --build 1019 --limit 5
Instructions
This skill orchestrates three sub-skills in order. Run each one by invoking it as a slash command.
Step 1: Gather failures
Run /gather-failures with the same arguments passed to this skill:
/gather-failures <job-path> --build <number> [--limit N] [--type TYPE]
This fetches Jenkins data, syncs repos, resolves test and product code, and saves a context file. Note the context file path from the output.
Step 2: Search for known fixes
Run /search-fixes with the context file path before analysis:
/search-fixes <context-file-path>
This searches the knowledge base for previously verified fixes that match the current failures. Note any known fixes found — they will be passed as context to the analysis step.
Step 3: Analyze failures (with known fixes as context)
Run /analyze-failures with the context file path from step 1:
/analyze-failures <context-file-path>
Important: If Step 2 found known fixes in the knowledge base, include them as additional context when analyzing. When writing the report, reference past fixes that match — use them to inform and improve the suggested fixes. For example, if a past fix shows that a similar NoSuchElementException was resolved by updating an airgun locator, recommend the same pattern.
This reads the context file and all referenced source files, groups failures by root cause, and writes a detailed report with suggested fixes informed by the knowledge base.
Step 4: Final summary
After all three steps complete, print a final summary:
- Total failures and root cause groups
- How many are test fixes vs product bugs
- Any known fixes found in the knowledge base and how they informed the report
- Path to the report file