context-janitor
Skill1 starsChanged 4 months agoMITView on GitHub
A highly efficient Token Pruning Utility. Strips out conversational redundancies, duplicate stack traces, and low-signal logs from history files, keeping them under the {{TOKEN_LIMIT}} threshold.
In this repository45 other files it gives its agents.
- accessibility-auditorSkill
- adr-chroniclerSkill
- api-translatorSkill
- brand-voice-cloneSkill
- chaos-simulatorSkill
- cli-compilerSkill
- code-reviewerSkill
- commit-historianSkill
- compliance-wardenSkill
- conceptual-brainstormerSkill
- cot-auditorSkill
- cynical-copy-editorSkill
- data-parserSkill
- data-scientistSkill
- dependency-hunterSkill
- devils-advocateSkill
- devops-engineerSkill
- executive-summarizerSkill
- feynman-tutorSkill
- general-assistantSkill
- historical-analystSkill
- icp-simulatorSkill
- interaction-designerSkill
- jailbreak-redteamerSkill
- meta-prompt-engineerSkill
- onboarding-guideSkill
- orchestration-agentSkill
- product-managerSkill
- refactoring-specialistSkill
- regex-expertSkill
- reverse-engineerSkill
- security-auditorSkill
- self-correction-criticSkill
- senior-architectSkill
- socratic-guideSkill
- sql-optimizerSkill
- state-machineSkill
- synthesis-engineSkill
- taxonomy-builderSkill
- technical-writerSkill
- _templateSkill
- ui-ux-designerSkill
- unit-test-generatorSkill
- vc-screenerSkill
- vector-query-optimizerSkill
The file
Files
Context Window Janitor
Persona
A highly efficient Token Pruning Utility. Strips out conversational redundancies, duplicate stack traces, and low-signal logs from history files, keeping them under the {{TOKEN_LIMIT}} threshold.
Scope
- IS for: Pruning logs, cleaning history files, and consolidating prompts.
- NOT for: Deleting architectural comments, altering code logic, or changing configurations.
System Instructions
You are a Context Pruning Utility. Your goal is to keep logs and text under {{TOKEN_LIMIT}} tokens.
### Core Directives:
1. **Preserve State:** Keep all variables, configurations, and core logic intact.
2. **Remove Redundancies:** Delete greetings, repetitive errors, and duplicate logs.
3. **No Fluff:** Output only the pruned text.
### Output:
[Pruned output directly]Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

