Agent skill
fix
(ePost) Use when user says "fix", "broken", "error", "failing", "this doesn't work", or pastes an error message — identifies error type and applies platform-appropriate fix
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/fix-klara-copilot-epost-agent-kit
Metadata
Additional technical details for this skill
- argument hint
- [issue description]
SKILL.md
Fix — Unified Fix Command
Fix issues with automatic error type detection. Absorbs :fast, :test, :types, :logs into one auto-detecting command.
Step 0 — Flag Override
If $ARGUMENTS starts with --ci: skip auto-detection, load references/ci-mode.md and execute. Remaining args are the issue description.
If $ARGUMENTS starts with --deep: skip auto-detection, load references/deep-mode.md and execute.
If $ARGUMENTS starts with --ui: skip auto-detection, load references/ui-mode.md and execute inline (main context — do NOT fork).
If $ARGUMENTS starts with --a11y: skip auto-detection, load references/a11y-mode.md and execute inline (main context — do NOT fork).
Otherwise: continue to Error Type Auto-Detection.
Dispatch Routing
| Flag | Execution | Specialist |
|---|---|---|
--ui |
Inline in main context | Main context dispatches epost-muji via Agent tool |
--a11y |
Inline in main context | Main context dispatches epost-a11y-specialist via Agent tool |
--deep |
Dispatch to epost-debugger via Agent tool | epost-debugger runs deep investigation |
--ci |
Dispatch to epost-debugger via Agent tool | epost-debugger handles CI log analysis |
| (auto-detect) | Dispatch to epost-debugger via Agent tool | epost-debugger handles general debugging |
Aspect Files
| File | Purpose |
|---|---|
references/deep-mode.md |
Systematic deep fix with full investigation |
references/ci-mode.md |
Fix CI pipeline failures |
references/ui-mode.md |
Fix UI component findings from known-findings DB |
references/a11y-mode.md |
Fix accessibility findings from known-findings.json |
Error Type Auto-Detection
Dispatch to epost-debugger via Agent tool with the full issue context before proceeding. epost-debugger will handle detection, investigation, and fixing. The modes below are instructions for epost-debugger.
Before fixing, detect the error type from context:
1. TypeScript Errors (formerly types subcommand)
Detection: Web platform detected AND tsconfig.json exists
Action: Run tsc --noEmit → fix all type errors → repeat until clean (zero errors)
Rules: NEVER use any type. Use proper narrowing, generics, utility types. No @ts-ignore.
2. Test Failures (formerly test subcommand → use /fix-deep for systematic)
Detection: User mentions "test" OR recent test runner output shows "FAIL"/"ERROR" Action: Run test suite → analyze failures → fix production code (not tests) → re-run until green Rules: Do NOT comment out or skip tests. Do NOT change assertions. Fix root causes.
3. Log-Based (formerly logs subcommand)
Detection: User provides log file path OR ./logs.txt exists
Action: Read log file → grep errors (last 30 lines) → locate in codebase → fix → verify logs clean
Rules: Fix ALL logged errors, not just the first one. Set up log piping if missing.
4. Quick Fix (default mode, formerly fast subcommand)
Detection: None of the above matched Action: Quick diagnosis → minimal correct change → verify (typecheck, tests, build) → add regression test Rules: Fix root causes, not symptoms. Keep changes minimal.
Platform Detection
Same as /cook — detect from changed files or $ARGUMENTS platform hint.
Explicit Overrides
For cases where auto-detection isn't enough, use flags:
/fix --deep— full systematic investigation with documentation/fix --ci— CI pipeline debugging (reads CI logs, reproduces locally)/fix --ui— UI component findings from.epost-data/ui/known-findings.json(delegates to epost-muji)/fix --a11y— accessibility findings from.epost-data/a11y/known-findings.json
$ARGUMENTS
IMPORTANT: Analyze the skills catalog and activate needed skills.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?