メニュー
Use when optimizing Agent Skills for lower token usage, creating compact companion skills, splitting skill instructions into references, extracting deterministic workflow scripts, or adding eval-backed skill compression
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Use when initializing Universal Token Killer schema artifacts for a project, discovering registered Copilot tools, seeding .utk output schemas, or narrowing schema generation to specific tool ids with descriptions or sample outputs
Use when installing, operating, recovering, or extending Universal Token Killer for GitHub Copilot tool-hook mediation, .utk artifacts, serializer configuration, schema routing, compact tool-output recovery, or RTK-parity token optimization
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
Use when configuring, testing, or documenting UTK model-proxy behavior for GitHub Copilot context optimization.
| name | detoks-skill |
| description | Use when optimizing Agent Skills for lower token usage, creating compact companion skills, splitting skill instructions into references, extracting deterministic workflow scripts, or adding eval-backed skill compression |
Make token-optimized companion Agent Skills. Preserve original. Generate compact root, move detail to references/, extract deterministic work to scripts/, validate with AgentEvals.
references/workflow.md: end-to-end source inspect, companion generation, token budget report.references/compression-strategies.md: TOC split, detok/caveman pass, preservation rules.references/script-extraction.md: deterministic workflow detection and TS script extraction.references/evals.md: AgentEvals EVAL.yaml contract and acceptance checks.scripts/analyze-skill.ts: token hotspots + deterministic candidates.scripts/render-optimized-skill.ts: write companion skill without mutating source.scripts/validate-optimized-skill.ts: frontmatter, references, token ratio, preservation checks.scripts/evolve-candidates.ts: optimize candidates through real Microsoft Trace (trace-opt) or Agent Lightning (agentlightning) backend.scripts/optimize-agent-frontmatter.ts: compact frontmatter context, preserve declarations.references/; put deterministic repeated steps in scripts/.detoks-prompt CLI for reference prose compression when available.evals/EVAL.yaml as quality gate before reporting success.