self-managed-context
Gives the model read-write control of its own context window: context as an editable file, eviction, harness safety invariants and prefix-cache costs.
- Kind
- Skill
- Works with
- Any agent
- Source
- Agent Skills for Context Engineering @ 9ff0c99108b3
- Path
- skills/self-managed-context/SKILL.md
- Jobs
- Context management, Guardrails
- Topics
- LLM apps & RAG, Agents & MCP, Machine learning
Credit
- Made by
- muratcankoylan
- Original
- github.com/muratcankoylan/Agent-Skills-for-Context-Engineerin…
- License
- MIT
Listed by Skill Manifest; not made by us.
InstallInstall it from a plugin marketplace.
/plugin install context-engineering@context-engineering-marketplace
Install factsScanned 2 Nov 2026
- Adds
- .claude/skills/self-managed-context/ (the skill folder)
- 2 file(s) the skill refers to
- Runs scripts
- Yes
- shell block in the text
- Uses the network
- None found
- Needs secrets
- None found
- Source revision
- 9ff0c99108b3
- Licence
- MIT
- Last activity
- 12 Oct 2026
- Review
- Reviewed
Found by a static scan of the stored text; nothing is run. “None found” is not a guarantee, so read what you install before you run it.
- Only the main file is read; other files in the folder are not scanned
# Self-Managed Context This skill covers agents that manage their own context window: the model, not the harness, decides what stays in the live context, what is compacted, what is evicted, and what is rewritten in place. The reference implementation is Context Language Models (CLMs), which mirror the editable part of the conversation into a file that the model edits with ordinary code tools, then re-parse that file into the next prompt. Applied zero-shot to existing models on a shared agent backbone, this matched or beat harness-scheduled and action-based context management on accuracy and compute across deep-research, terminal-coding, and multi-hour repository-optimization tasks (claim-self-managed-context-zero-shot-results, claim-self-managed-context-long-horizon-results). …