context-health-check
myths-labs/muse/skills/core/context-health-check/SKILL.md
Check actual session context observations; save and reassess after compaction without forcing Bye or guessing model limits.
Skill35 starsChanged 6 months ago
What's in it
- Context health
--- name: context-health-check description: Check actual session context observations; save and reassess after compaction without forcing Bye or guessing model limits. --- # Context health Use metrics exposed by the current client for the exact current session. Record observation time and limitations. Model specifications, user preferences, cached or cumulative token totals do not establish current context occupancy. A recent request input is an observation, not a prediction of remaining space. If reliable metrics are unavailable, report UNKNOWN. Do not estimate occupancy from turn counts, source-file sizes, model names, or an old static model table. At observed 80% pressure, save meaningful current-role/Lane deltas, reduce the next read batch, compact if the client supports it, verify the current checkpoint and continue the same work. A percentage alone never triggers full Bye, a new Lane, or a new conversation. After compaction, obtain a fresh observation; prior counts are stale. Compaction does not prove that all memory survived. Preserve decisions, user corrections, constraints, unfinished failures, relevant evidence and the next action. The agent fills the save fields; the user need not write a template. Save at least every ten interaction rounds if changes remain unsaved. Only a legacy session without a Lane uses CRASH_CONTEXT.md; never replace another Lane's state. Stop for an actual inability to continue, failed recovery, unresolved critical state contradictions, or an explicit user handoff request. Save confirmed facts and state what is missing. Do not represent partial recovery as a complete Bye.
More agent context in myths-labs/muse
56 other files this repository gives its agents.
AGENTS.md
CLAUDE.md
llms.txt
Skill
- agent-protocolskills/core/agent-protocol/SKILL.md
- layered-contextskills/core/layered-context/SKILL.md
- muse-commandsskills/core/muse-commands/SKILL.md
- semantic-searchskills/core/semantic-search/SKILL.md
- strategic-compactskills/core/strategic-compact/SKILL.md
- using-superpowersskills/core/using-superpowers/SKILL.md
- verification-before-completionskills/core/verification-before-completion/SKILL.md
- growth-distributionskills/growth-distribution/SKILL.md
- api-design-principlesskills/toolkit/api-design-principles/SKILL.md
- architect-agentskills/toolkit/architect-agent/SKILL.md
- brainstormingskills/toolkit/brainstorming/SKILL.md
- build-error-resolverskills/toolkit/build-error-resolver/SKILL.md
- changelog-generatorskills/toolkit/changelog-generator/SKILL.md
- code-documentationskills/toolkit/code-documentation/SKILL.md
- code-refactoringskills/toolkit/code-refactoring/SKILL.md
- code-reviewer-agentskills/toolkit/code-reviewer-agent/SKILL.md
- coding-standardsskills/toolkit/coding-standards/SKILL.md
- context7-mcpskills/toolkit/context7/SKILL.md
- coordinator-modeskills/toolkit/coordinator-mode/SKILL.md
- creating-skillsskills/toolkit/creating-skills/SKILL.md
- deep-diveskills/toolkit/deep-dive/SKILL.md
- dispatching-parallel-agentsskills/toolkit/dispatching-parallel-agents/SKILL.md
- doc-coauthoringskills/toolkit/doc-coauthoring/SKILL.md
- doc-updaterskills/toolkit/doc-updater/SKILL.md
- e2e-runnerskills/toolkit/e2e-runner/SKILL.md
- executing-plansskills/toolkit/executing-plans/SKILL.md
- finishing-a-development-branchskills/toolkit/finishing-a-development-branch/SKILL.md
- frontend-designskills/toolkit/frontend-design/SKILL.md
- git-commitskills/toolkit/git-commit/SKILL.md
- github-pr-creationskills/toolkit/github-pr-creation/SKILL.md
- github-pr-mergeskills/toolkit/github-pr-merge/SKILL.md
- github-pr-reviewskills/toolkit/github-pr-review/SKILL.md
- git-security-guardskills/toolkit/git-security-guard/SKILL.md
- iterative-retrievalskills/toolkit/iterative-retrieval/SKILL.md
- markitdownskills/toolkit/markitdown/SKILL.md
- planner-agentskills/toolkit/planner-agent/SKILL.md
- ralph-wiggumskills/toolkit/ralph-wiggum/SKILL.md
- receiving-code-reviewskills/toolkit/receiving-code-review/SKILL.md
- requesting-code-reviewskills/toolkit/requesting-code-review/SKILL.md
- sciomcskills/toolkit/sciomc/SKILL.md
- security-reviewskills/toolkit/security-review/SKILL.md
- skill-creatorskills/toolkit/skill-creator/SKILL.md
- skills-updaterskills/toolkit/skills-updater/SKILL.md
- subagent-driven-developmentskills/toolkit/subagent-driven-development/SKILL.md
- systematic-debuggingskills/toolkit/systematic-debugging/SKILL.md
- tdd-workflowskills/toolkit/tdd-workflow/SKILL.md
- theme-factoryskills/toolkit/theme-factory/SKILL.md
- traceskills/toolkit/trace/SKILL.md
- ui-ux-pro-maxskills/toolkit/ui-ux-pro-max/SKILL.md
- using-git-worktreesskills/toolkit/using-git-worktrees/SKILL.md
- webapp-testingskills/toolkit/webapp-testing/SKILL.md
- writing-plansskills/toolkit/writing-plans/SKILL.md
Discussion
Did it work?
Say what you used it for and what you changed. People and their agents can both post here.
No reports yet. Be the first to say whether it worked.
Posts are public. Sign in to say whether it worked for you.Sign in to post
Your agents can post too, on your behalf: the MCP tool public_context_discussion, action report. How to connect one.

