| name | detoks |
| description | Use when compressing prompts, agent skills, AGENTS.md files, GitHub Copilot custom agents, subagent definitions, or bulky artifacts before sending text to LLM context; includes detoks-prompt CLI/MCP use, skill consolidation, AGENTS.md memory/skill extraction, and GHCP subagent/frontmatter orchestration |
Detoks
Orchestrate all detoks workflows through this skill. Keep SKILL.md as router only; load focused references before changing prompt, skill, AGENTS.md, or GitHub Copilot custom-agent content.
Route
- Prompt, reusable instruction, large prose artifact: read
references/detoks-prompt.md.
- Skill or old
detoks-skill content: read references/detoks-skill.md.
- AGENTS.md cleanup: read
references/detoks-agentsmd.md, then references/detoks-prompt.md.
- GitHub Copilot custom agent or subagent refactor: read
references/detoks-ghcp-subagent.md, then references/detoks-prompt.md.
- MCP tool details or rates: read
references/detok-mcp.md.
Core Rules
- Preserve raw source before compression or refactor.
- Use file/stdin CLI flow for large prompt text; avoid pasting bulky source into chat.
- Preserve code blocks, inline code, quoted requirements, frontmatter keys, tool names, file paths, links, and exact validation commands.
- Prefer references and extracted skills over one huge instruction file.
- State when conclusions depend on compressed text.