zeph-skills
SKILL.md parser, registry, embedding matcher, and hot-reload for Zeph.
Overview
Parses SKILL.md files (YAML frontmatter + markdown body) from the skills/ directory, maintains an in-memory registry with hot-reload support, and formats selected skills into LLM system prompts. Supports semantic matching via Qdrant embeddings and self-learning skill evolution with trust scoring.
Key modules
| Module | Description |
|---|---|
loader |
SKILL.md parser (YAML frontmatter + markdown) |
registry |
In-memory skill registry with hot-reload |
matcher |
Keyword-based skill matching |
qdrant_matcher |
Semantic skill matching via Qdrant |
evolution |
Self-learning skill generation and refinement |
trust |
SkillTrust — skill trust scoring; re-exports TrustLevel from zeph-tools |
watcher |
Filesystem watcher for skill hot-reload |
prompt |
Skill-to-prompt formatting (full, compact, auto modes via SkillPromptMode) |
manager |
SkillManager — install, remove, verify, and list external skills (~/.config/zeph/skills/) |
Re-exports: SkillError, SkillTrust, TrustLevel (from zeph-tools), compute_skill_hash
Prompt modes
The prompt_mode config option ([skills] section) controls how skills are serialized into the LLM system prompt:
| Mode | Description |
|---|---|
full |
Full XML format with complete skill body (default) |
compact |
Condensed XML with name, description, and trigger list only |
auto |
Selects compact when context budget is below threshold, full otherwise |
Installation
License
MIT