workflow-context-hygiene
Skill1 starsChanged 6 months agoMITView on GitHub
Keep the working set small and the plan always in focus.
In this repository33 other files it gives its agents.
- skill-process-gotchasSkill
- workflow-cache-disciplineSkill
- workflow-codegraphSkill
- workflow-conductorSkill
- workflow-context-condenserSkill
- workflow-criticSkill
- workflow-deadline-disciplineSkill
- workflow-deep-researchSkill
- workflow-done-definitionSkill
- workflow-edit-formatSkill
- workflow-effort-routingSkill
- workflow-escalateSkill
- workflow-execute-from-planSkill
- workflow-filesystem-offloadSkill
- workflow-learn-frameworkSkill
- workflow-long-running-processesSkill
- workflow-lspSkill
- workflow-memory-curationSkill
- workflow-plan-contractSkill
- workflow-plan-firstSkill
- workflow-planSkill
- workflow-quality-barSkill
- workflow-repomapSkill
- workflow-retrieve-before-actSkill
- workflow-spec-drivenSkill
- workflow-stuck-detect-resetSkill
- workflow-stuckSkill
- workflow-subagent-isolationSkill
- workflow-subtask-loopSkill
- workflow-task-finishSkill
- workflow-task-startSkill
- workflow-tool-selectionSkill
- workflow-verify-enhancedSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 12
Keep in active focus only: the current subtask, the code being changed now, relevant retrieved snippets (from codebase-rag — …
- Distinctive: line 16
Do NOT paste large file contents repeatedly.
- Very specific: line 18
After finishing a subtask, you do not need its working details anymore — let them compress.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
<!-- TRIGGERS WHEN: Keep the working set small and the plan always in focus. -->The model attends poorly to long context. Keep the relevant set small and pinned.
- PLAN.md is the source of truth, not memory. Re-read it at each subtask start. If your memory of the plan and PLAN.md disagree, PLAN.md wins.
- Keep in active focus only: the current subtask, the code being changed now, relevant retrieved
snippets (from codebase-rag —
mcp_hermes_max_codebase_rag_search_code/mcp_hermes_max_codebase_rag_get_symbol_context), and the rules. Everything else can be compressed/forgotten. - Do NOT paste large file contents repeatedly. Retrieve the specific function with codebase-rag when needed instead of holding whole files in context.
- After finishing a subtask, you do not need its working details anymore — let them compress.
The durable record is the git checkpoint (
mcp_hermes_max_checkpoint_checkpoint) + the knowledge-graph entry, not the chat history.
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.