| name | repo-cognition |
| description | ai-native repository cognition and context management framework for long development cycles, memory management, retrieval systems, multi-agent workflows, operational continuity, and context-aware repository governance. use when working with project context, handoffs, repository memory systems, context windows, token management, retrieval optimization, or ai-assisted software development workflows. |
Repository Cognition
This skill establishes operational rules and retrieval systems for AI-assisted repositories. This skill specifically is part of the Documentation.md Repository and should adhere to its rules and guidelines. But, can be incorperated into other repositories and systems in part.
The repository should be treated as operational memory shared between humans and AI systems.
The goal is not to remember everything. Reference - Context-Entropy, and the Docs/Context folders for more information on CWM, Entropy, Decay, and CTL.
The goal is to retrieve the right information at the right time while preserving:
- continuity
- clarity
- retrieval quality
- implementation awareness
- contextual understanding
Follow the principles established in:
- references/context-entropy.md
- references/context-window.md
- references/context-rules.md
- references/context-token-limits.md
Maintain awareness of:
- context drift
- context saturation
- context decay
- token usage
- retrieval quality
- active project knobs
- implementation cycles
Avoid:
- unnecessary retrieval
- duplicated summaries
- repository fragmentation
- noisy artifacts
- hallucinated implementation details
- excessive context ingestion
Prioritize:
- active implementation context
- retrieval clarity
- token efficiency
- repository continuity
- operational awareness
The repository is not static documentation and will be updated periodically based on user feedback and project changes.
It is active operational memory.
