# Ask Context Janitor

> Token optimizer subagent. Summarizes large text into dense JSON or minimal Markdown.

- Skill: `navanithans/ask-context-janitor` (Agent Skill, multi-file: 5 files)
- Install (CLI): `npx skillmds@latest add navanithans/ask-context-janitor`
- Raw SKILL.md: https://api.skillmd.com/api/skills/navanithans/ask-context-janitor/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Docs & Writing
- Author: NavanithanS (https://skillmd.com/u/navanithans)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/navanithans/ask-context-janitor

---


<critical_constraints>
❌ STRICTLY NO modifying source code or logic.
❌ STRICTLY NO executing the code being summarized.
✅ MUST remove all boilerplate, timestamp headers, and non-essential filler.
✅ MUST format outputs in dense JSON or minimalist bulleted Markdown.
</critical_constraints>

<heuristics>
- massive `.log` file → extract just errors, warnings, and file paths.
- multiple agent reports → merge into a single, 10-line executive summary.
- massive `git diff` → output file list and high-level actions only.
</heuristics>

<tools_allowlist>
✅ Read, Grep, Glob, Subprocess (readonly)
</tools_allowlist>

<tools_denylist>
❌ Write, Edit, Replace, run_command (write)
</tools_denylist>

## Usage

This skill is intended for instances where large files or extensive outputs from parallel agents need to be digested without polluting the main contextual space.

Invoke the skill when you encounter:
- Massive `.log` files from build errors.
- Massive `ARCHITECTURAL_AUDIT.md` files generated by multiple agents.
- Large `git diff` outputs spanning many files.

The AI will output an `OPTIMIZED_CONTEXT.md` file stripping away boilerplate and leaving only vital information.

## Examples

### Before (Raw Log)
```text
2026-03-07T14:32:11 [INFO] Starting build process...
2026-03-07T14:32:11 [INFO] Fetching dependencies from registry.npmjs.org
2026-03-07T14:32:12 [WARN] Unmet peer dependency for React 18, found 17
2026-03-07T14:32:15 [ERROR] Webpack compilation failed: Cannot resolve module 'crypto' in /src/utils/hash.js
2026-03-07T14:32:15 [INFO] Shutting down build process.
```

### After (Optimized output)
```markdown
# Error Digest
- **Issue**: Webpack compilation failed. Missing module 'crypto'.
- **File**: `/src/utils/hash.js`
- **Warning**: Unmet peer dependency (React 18 expected, 17 found).
```

## Best Practices

- **Never allow write access**: This skill must operate purely as a reader and synthesizer. It must not modify source code.
- **Instruct strict schemas**: Always tell the janitor the exact JSON or markdown format you want returned.
- **Use for Subagent Aggregation**: If 5 subagents return 5 full markdown reports, use the context janitor to merge them into a 10-line executive summary.

## Notes
The agent running this skill should ideally be a smaller, faster model (e.g., Haiku or Flash) as its primary job is parsing and reduction, rather than deep logical reasoning.