Benchmark Yylo
Prepare reviewed cases and frozen checklists, compare models and harnesses, and independently evaluate retained outputs with YYLO Benchmark's thin trusted-host runner.
Install
npx skills add yylo-dev/yylo-skills --skill benchmark-yylogit clone --depth 1 https://github.com/yylo-dev/yylo-skills.git /tmp/yylo-skills
mkdir -p ~/.claude/skills
cp -r /tmp/yylo-skills/skills/benchmark-yylo ~/.claude/skills/benchmark-yyloPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Run a Benchmark experiment ## Discover the actual runtime Inspect `yylo-benchmark --version` and `yylo-benchmark --help` (or delegated `yy benchmark --help`) before using commands. This guidance targets Benchmark 0.2.1's thin-runner/checklist contract, not retired 0.1.x. A source merge or skill update does not install, activate or publish that runtime. Stop on version mismatch; never silently substitute a checkout or upgrade an installation. The lifecycle is **case -> run -> evaluate -> report**, plus append-only `disqualify`. There is no automatic retry, repair, resume, combined judge/cost score or winner. Inspect subcommand help for exact options. Do not use the retired plan, recover, doctor, regrade or rejudge lifecycle or its plugin/governance machinery. Historical evidence remains readable with its original pinned implementation; do not migrate, delete or reinterpret it to match this runner. ## Prepare a reusable case For historical Ledger tasks read [references/historical-tasks.md](references/historical-tasks.md). `case draft --ledger-task TASK_ID` proposes the body and reference candidate, leaving the base unset. Review the requirements, original development range and answer-bearing paths; never assume the completion commit's parent is the base. Drafting is not automatic historical reconstruction or approval. For historical tasks, arbitrary coding prompts or workflows, prepare a reviewed case once and reuse it:
Details
- Skill name
benchmark-yylo- Repository
- yylo-dev/yylo-skills
- Path
skills/benchmark-yylo/SKILL.md- License
- MIT
- Repo updated
- 17 October 2026
- Listed in
- karanb192/awesome-claude-skills
FAQ
What does the Benchmark Yylo skill do?
Prepare reviewed cases and frozen checklists, compare models and harnesses, and independently evaluate retained outputs with YYLO Benchmark's thin trusted-host runner.
How do I install the Benchmark Yylo skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add yylo-dev/yylo-skills --skill benchmark-yylo. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Benchmark Yylo popular?
It lives in yylo-dev/yylo-skills, which has 1 GitHub stars and 1 forks, and is included in 1 curated skill list.
Related skills
Cohesivity
1cohesivity-org/cohesivity-plugin
Backend and infra for a project via Cohesivity (cohesivity.ai). Provisions Postgres, hosting and deploys, auth and social login, realtime websockets, an agent-native email inbox, object and vector storage, Redis, cron, and AI model APIs (OpenAI, Anthropic, Deepgram, Exa) through one HTTP API. Use when a valid .cohesivity file exists, a Cohesivity MCP is available, the user names Cohesivity, or the project needs a backend or any of these services and no competing provider was chosen.
Privacyscrubber Guard
3moxno/privacyscrubber-mcp
Pre-ingestion zero-trust PII, secret, and regulated data sanitization skill for AI coding agents (Claude Code, Cursor, Codex, Gemini CLI, Windsurf) via @privacyscrubber/mcp-server. Intercepts files, shell commands, and prompts in local RAM with 30 specialized industry profiles and SHA-256 compliance receipts.
AI Meeting
1bin1874/ai-meeting-skill
Run structured AI meetings for plans, product ideas, technical designs, business decisions, feature proposals, and strategy choices. Use when the user wants an AI meeting, AI roundtable, multi-agent discussion, debate, proposal review, plan review, decision review, or wants Codex, Claude Code, qoderclicn, and other CLI agents to analyze a proposal across multiple expert roles, challenge assumptions, preserve per-agent sessions, and produce a final decision report with provenance. 也适用于中文场景：方案评审、多 Agent 讨论、技术路线评审、商业决策评审。
AI Tell Detector
3aragossa/ai-tell-detector
Audits a finished draft for the patterns that make text read as AI-generated — rhetorical symmetry, generic filler, uniform rhythm, over-resolved story arcs, missing human roughness, fabricated personal experience — and flags each one with a fix. Use before publishing anything, or when a draft "feels AI" but you can't say why.
Crosscheck
1moveju112/skill_verify
Claude-hosted Claude↔Codex crosschecks of improvement ideas, plans, and completed changes. Automatically selects independent proposals, draft critique, or mutual review and allows candid rejection. Use for explicit requests such as "크로스체크", "코덱스랑 핑퐁", "둘이 의견 취합해", "cross-check with codex", or "/crosscheck". Discussing or editing this skill does not authorize another model call.
AI Code Review Triage
3laulauraluo/ai-code-review-triage
Review pull requests or diffs written by AI coding agents (Claude Code, Codex, Copilot, Cursor) when reading every line no longer scales, or when the code looks clean but you don't trust it yet. Splits "reading the code" into three separate jobs (catch bugs, align context, own the sign-off) and runs a different process for each. Use for requests like "review this AI-generated PR", "can this diff be merged", "too many agent PRs to review", "check this before I approve", or "what should I verify before signing off". For a plain security/performance checklist use a general code-review skill inste
Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.