Menu
Generate semantic embeddings for a repository's code symbols
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Use when the user wants their Kimi CLI agent to self-improve from past usage, asks about a nightly/offline "sleep" or "dream" cycle, memory/skill consolidation, or says things like "make my agent better the more I use it", "review my past sessions", "learn my preferences", "consolidate what you learned", "run the sleep cycle", or wants to schedule offline self-optimization.
Generate a structured knowledge report for a repository
Search across all registered repositories using hybrid (vector + keyword) search
| name | embed-repo |
| version | 1.0.0 |
| description | Generate semantic embeddings for a repository's code symbols |
| author | devbase-team |
| tags | ["embedding","semantic-search","indexing"] |
| skill_type | builtin |
| inputs | [{"name":"repo_id","type":"string","description":"Target repository ID in devbase","required":true},{"name":"device","type":"string","description":"Device for embedding model (cpu, cuda, auto)","default":"auto"}] |
| outputs | [{"name":"status","type":"string","description":"Completion status"}] |
This skill invokes the local embedding provider (examples/embedding-provider/local.py)
to generate 384-dimensional vectors for all code symbols in the specified repository.
devbase skill run embed-repo --arg repo_id=devbase --arg device=cuda
The entry script locates local.py relative to the devbase workspace and runs it
with the provided --repo-id and --device arguments.