sniper-check
npx machina-cli add skill fusengine/agents/sniper-check --openclawTarget: $ARGUMENTS
Sniper Check
Overview
Quick code quality validation using the sniper agent in an isolated forked context. Executes the full 6-phase workflow without polluting the main conversation context.
| Feature | Detail |
|---|---|
| Context | Forked (isolated sub-agent) |
| Agent | sniper (Sonnet) |
| Phases | 6-phase code-quality workflow |
| Result | Only final report returns to parent |
When to Use
| Scenario | Use |
|---|---|
| After code modifications | /sniper-check src/ |
| Validate specific file | /sniper-check path/to/file.ts |
| Full project check | /sniper-check . |
| After refactoring | /sniper-check src/components/ |
Workflow
Execute the mandatory 6-phase code-quality workflow:
- PHASE 1+2 (PARALLEL): Launch both in parallel:
explore-codebase(Haiku) → Architecture discoveryresearch-expert(Sonnet) → Documentation verification
- PHASE 3: Grep all usages → Impact analysis
- PHASE 4: Run linters → Detect errors
- PHASE 5: Apply corrections → Minimal changes
- PHASE 6: Re-run linters → Zero errors
CRITICAL: Phases 1+2 must run in PARALLEL (two Task calls in one message).
Critical Rules
| Rule | Reason |
|---|---|
| Never skip phases 1+2 | Documentation-backed fixes only |
| Always run phases in order | Dependencies between phases |
| Zero linter errors | Non-negotiable exit criteria |
| Minimal changes only | Smallest fix necessary |
Report Format
Return a validation report with:
- Architecture summary (from explore-codebase)
- Documentation references (from research-expert)
- Impact analysis table (usages, risk level)
- Errors fixed (critical, high, medium, low)
- SOLID compliance status
- Final linter status (must be ZERO errors)
Source
git clone https://github.com/fusengine/agents/blob/main/plugins/ai-pilot/skills/sniper-check/SKILL.mdView on GitHub Overview
sniper-check validates code quality after modifications by running the sniper agent in an isolated forked context. It executes the full 6-phase workflow and returns only the final report to the parent, keeping the main conversation context clean.
How This Skill Works
The tool forks a sub-agent and runs a mandatory 6-phase code-quality workflow. Phases 1 and 2 execute in parallel (explore-codebase and research-expert), followed by Phase 3 impact analysis, Phase 4 linting, Phase 5 minimal corrections, and Phase 6 re-run of linters until zero errors. The final report is returned to the parent while the fork remains isolated.
When to Use It
- After code modifications
- Validate a specific file
- Full project check
- After refactoring
- Before PR or release as part of CI checks
Quick Start
- Step 1: Run the command with a target path, e.g., /sniper-check src/ or /sniper-check path/to/file.ts
- Step 2: Wait for the 6-phase workflow to complete and review the final report returned to the parent
- Step 3: If any issues are found, apply minimal changes and re-run until the linter reports zero errors
Best Practices
- Never skip phases 1+2; rely on documentation-backed fixes only
- Always run phases in order to respect dependencies
- Zero linter errors is the non-negotiable exit criterion
- Minimal changes only; apply the smallest fix necessary
- Use the forked context to prevent polluting the main conversation
Example Use Cases
- After modifying a component, run /sniper-check src/ to validate the changes.
- Validate a specific file like path/to/file.ts to ensure no regressions were introduced.
- Run a full project check with /sniper-check . before a release.
- After refactoring a module, use sniper-check to verify architecture and documentation references.
- Integrate sniper-check into CI to auto-validate PRs and surface a final report with zero errors.
