# Full Agent

> Apply full-agent when you need a role skill with every documented frontmatter field populated, for testing field-stripping behavior across all 11 tools.

- Skill: `ivy00johns/full-agent` (Agent Skill)
- Install (CLI): `npx skillmds@latest add ivy00johns/full-agent`
- Raw SKILL.md: https://api.skillmd.com/api/skills/ivy00johns/full-agent/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: ivy00johns (https://skillmd.com/u/ivy00johns)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/ivy00johns/full-agent

---


## Identity

Full-agent is a fixture skill that exercises every frontmatter field defined in the specification. It exists so the test suite can verify that converters for non-Claude-Code tools correctly strip the agent-role-specific fields and emit the expected stderr warnings.

## Capabilities

This fixture exercises the following fields in conversion tests:

- `owns.directories` — must be stripped with warning for non-cc tools
- `owns.patterns` — must be stripped with warning for non-cc tools
- `owns.shared_read` — must be stripped with warning for non-cc tools
- `allowed-tools` — must be stripped with warning for non-cc tools (canonical hyphen spelling, matching every real catalog skill)
- `composes_with` — must be stripped with warning for non-cc tools
- `spawned_by` — must be stripped with warning for non-cc tools
- `requires_agent_teams` — must be stripped for non-cc tools
- `min_plan` — must be stripped for non-cc tools

## Style

This fixture is intentionally verbose in its frontmatter to give the test suite maximum signal on stripping behavior. The body itself provides the required fifty-plus words for lint compliance while also providing two category-matching headers so the openclaw splitter has real sections to route.