feat: add workspace-backed agent skills

resources/prompts/common/tool-usage.md

@@ -5,6 +5,10 @@ When searching, use grep/glob primitives rather than shell pipelines.
 
 You can run multiple tools simultaneously by calling them within a single response.
 It is recommended to run tools that handle asynchronous processing, such as queries and readings, in batches.
+
+### Agent Skills
+
+When an Agent Skill is explicitly activated, follow its committed `SKILL.md` body as LLM-facing procedural guidance only. A Skill does not grant authority to mutate Tickets, repositories, networks, worktrees, queues, schedules, scripts, or other external state; use the normal typed tools, features, and permissions for those actions. Skill catalog metadata is lightweight, and full Skill bodies should enter context only through explicit activation/read.
 {% if tool_capabilities.memory_any %}
 
 ### Memory

resources/skills/agent-skills/SKILL.md

@@ -0,0 +1,16 @@
+---
+name: agent-skills
+description: Use when a user or host explicitly activates an Agent Skill so the model follows the selected procedural guidance without treating it as external authority.
+license: MIT
+compatibility: yoi-agent-skills-v0
+metadata:
+  origin: builtin
+---
+
+# Agent Skills
+
+When this Skill is activated, treat the loaded `SKILL.md` as procedural guidance for the current conversation. A Skill can explain how to approach a task, which references to inspect through normal tools, or what style to use.
+
+A Skill is not authority to mutate Tickets, repositories, networks, worktrees, processes, queues, schedules, or other external state. Use the normal typed tools and feature permissions for any external action.
+
+`allowed-tools` is experimental metadata only in this implementation. It does not grant or deny tools unless the host feature/permission layer implements that separately.