skills/sentry-analysis/SKILL.md
Analyze Sentry error logs, breadcrumbs, and codebase context to diagnose and explain the root cause of issues.
npx skillsauth add preetamnath/agent-skills sentry-analysisInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
You are an expert debugging assistant. Your task is to investigate Sentry errors and determine their root cause, or all probable root causes, based on provided logs and codebase context.
Ingest Context:
AskUserQuestion tool with options: "Provide additional context (user ID, deploy SHA, repro steps)", "Proceed with available data". Recommended: "Provide additional context".Codebase Investigation:
Analyze:
AskUserQuestion tool to present the candidates with your best-guess ranking and ask which to investigate further before finalizing the report.Report Findings:
AskUserQuestion tool with options: "Agree with triage — done", "Adjust triage level", "Hand off to fix skill". Recommended: based on triage level ("Hand off to fix skill" for Critical, "Agree with triage — done" for Edge Case/Noise).Constraints:
documentation
Collapse a multi-clause instruction into one positive line of trigger + action. TRIGGER when: user says 'tighten this rule', 'make this leaner', 'make this simpler' in a skill, CLAUDE.md, agent prompt, or style guide.
documentation
File-level tightening pass on an instruction file (CLAUDE.md, skill, agent prompt, style guide) using `tighten-instruction` as the lens. TRIGGER when: user says 'tighten/simplify this file/skill/CLAUDE.md', 'cut this down'; user points at a verbose instruction file and wants it leaner.
testing
Anchored second-opinion on one concrete proposal: dispatch a subagent to rate the fix, generate ranked alternatives, and flag blind spots, then synthesize back. TRIGGER when: user says 'second opinion', 'rate my fix', 'weigh in on my approach', 'what alternatives am I missing', or wants their candidate edit/decision evaluated against alternatives. SKIP when: multiple decisions on a larger artifact — use `panel-review`.
development
Multi-reviewer panel on N focused questions about a near-final artifact (plan, design, code, prose). R0 (you) plus two parallel reviewer subagents, per-question table with disagreement preserved, walk decisions one at a time. TRIGGER when: user says 'panel review', 'multi-agent review'; user has a mostly-done artifact and focused micro-decisions to validate. SKIP when: only one proposal under review — use `second-opinion`.