docs(coding-agent): note that models often need prompting to use skills

This commit is contained in:
Mario Zechner 2026-01-25 20:54:10 +01:00
parent 77469e1599
commit b29de1c73d

View file

@ -58,7 +58,7 @@ For project-level Claude Code skills, add to `.pi/settings.json`:
1. At startup, pi scans skill locations and extracts names and descriptions 1. At startup, pi scans skill locations and extracts names and descriptions
2. The system prompt includes available skills in XML format per the [specification](https://agentskills.io/integrate-skills) 2. The system prompt includes available skills in XML format per the [specification](https://agentskills.io/integrate-skills)
3. When a task matches, the agent uses `read` to load the full SKILL.md 3. When a task matches, the agent uses `read` to load the full SKILL.md (ideally; models often need prompting)
4. The agent follows the instructions, using relative paths to reference scripts and assets 4. The agent follows the instructions, using relative paths to reference scripts and assets
This is progressive disclosure: only descriptions are always in context, full instructions load on-demand. This is progressive disclosure: only descriptions are always in context, full instructions load on-demand.