experiment-coder
Implement, deploy, monitor, sync, and debug experiments from the STATE.md Runs table.
An agent is a specialist Claude hands a whole job to, with its own tools and its own context.
2,200 agents across 245 plugins.
Implement, deploy, monitor, sync, and debug experiments from the STATE.md Runs table.
Review an experiment workspace to top-conference standards, then write the final verdict and next phase.
Analyze experiment results, respond to audits and reviews, update STATE.md, and plan the next experiment round.
The ONLY agent allowed to write files. Used exclusively by the fix skill (the /claude-seo-ai:fix command) AFTER the user has confirmed the diffs. Applies…
Read-only technical SEO specialist. Use proactively during an audit to analyze crawlability, indexability, rendering, Core Web Vitals, mobile-friendliness,…
Use when a Compound V manifest (or a plan with a verified Partition Map) is ready to execute and you want to offload the batched, manifest-driven,…
Use when a Compound V manifest (or a plan with a Partition Map) is ready and you need to verify its partition is genuinely disjoint and its invariants hold…
Use to run Compound V's three-pass Review Gate. Pass 1 SPEC — the change matches the task spec and the manifest's feature-level acceptance_criteria. Pass 2…
Creative problem-solving across the full complexity spectrum. Finds solutions nobody proposed.
Discovers who is affected and diagnoses what they feel. Actor discovery and emotional granularity as diagnostic tools.
Creates clear, actionable documentation from synthesized analysis. Invisible framework.
Execute ENCODE DNase-seq pipeline from FASTQ to hotspots and footprints using BWA, Hotspot2, and HINT-ATAC
Execute ENCODE Hi-C pipeline from FASTQ to contact matrices and loop calls using BWA, pairtools, Juicer, and HiCCUPS
Execute ENCODE RNA-seq pipeline from FASTQ to gene quantification using STAR 2-pass alignment and RSEM/Kallisto
事後分析代理 —— 分析盲測比較結果以理解為什麼贏家勝出,並生成改進建議。 也用於分析基準測試結果,發現彙總指標可能隱藏的模式和異常。 使用範例: - "分析為什麼版本 A 的技能比版本 B 更好" - "審查基準測試結果並找出模式" - "分析測試執行結果並生成改進建議"
設計功能架構和實作藍圖。分析需求、評估多種實作方案、考慮權衡, 並推薦最適合程式碼庫的解決方案。 使用時機範例: - "設計新的使用者通知系統" - "規劃支付整合架構" - "設計可擴展的檔案上傳系統" - "架構新的 API 端點結構"
深入分析現有程式碼庫功能,透過追蹤執行路徑、映射架構層次、理解模式和抽象化,以及記錄相依性。 使用時機範例: - "分析使用者驗證功能的實作方式" - "了解支付處理流程如何運作" - "探索 API 路由系統的架構" - "追蹤資料庫查詢的執行路徑"
Final trading decision maker with paper trading execution. Use after gathering analysis from other agents to make EXECUTE/WAIT/REJECT decisions.
Portfolio risk management, volatility analysis, and market microstructure. Use for VaR calculations, correlation studies, orderbook depth, position sizing, and…
Self-evolving platform agent. Researches APIs, generates new MCP servers, agents, and skills following existing patterns. Used by the /create skill.
Produces images and videos for social posts with human-in-the-loop approval at every creative stage.
Reviews generated creative for brand consistency, visual quality, copy accuracy, and platform compliance before approval queue.
Final quality gate for beast-plan. Aggregates all actor feedback, scores comprehensively, and issues APPROVED/REVISE/REJECT verdict.
Verifies dead code findings from knip/grep. Reads actual source files to filter false positives. Cites file:line for every finding.
Independent verification agent. Takes FINAL-PLAN.md and independently verifies every acceptance criterion was met. No access to executor output.
Independent review of work someone else finished - a diff, a feature, a migration. Sees only the change and the criteria, never the reasoning that produced it.…
Use when: /map command, finding skills/agents, understanding ecosystem layout. Do NOT use for: code generation, debugging, file editing.
Use when: before the lead reports a root-cause conclusion, a 'done/verified' claim, an irreversible action about to run (commit/deploy/rm/push), or a 2nd-time…
Use when: checking for Claude Code updates (/watch command), detecting breaking changes in our plugins, monitoring community feedback (/watch --pulse). Do NOT…
Use this subagent to map styles and variables in Figma, Sketch, MasterGo, Pixso, Mokou, or Mockup to existing design tokens, theme variables, and style…
Front-end warehouse document synchronization subagent: synchronizes README, runtime docs, project structure, capability tables, and report descriptions from…
Front-end end-to-end testing specialist: writing and maintaining key user journeys, executing Playwright/Cypress, managing unstable use cases, managing…
Maintainer-only advisory audit of GAIA instruction-prose under .claude/skills/**/*.md for gratuitous complexity: prose too long, too deeply nested, too…
Maintainer-only audit of framework bash and the bats suites guarding it: quoting/portability correctness, a shellcheck oracle, and conditional hook-contract…
- `@conform-to/zod` must **always** be imported from the `/v4` subpath. Flag any import from the bare `@conform-to/zod` package as **Critical**. The default…
Independent evidence-based verification. Use after non-trivial work to check the result against the plan - run tests, lint, and type checks, verify checklist…
Identifies visual opportunities in content, generates data charts from verified research, optionally generates AI images (feature images, contextual…
Universal draft-vs-research hallucination and citation-integrity audit — runs on ALL content types, not just scientific or technical pieces. Diffs every…
Optimizes content structure for readability and engagement, and catches grammar, spelling, and formatting errors.
You are the **Final Solution Synthesizer** in a Software Engineering Chain of Verification (SE-CoVe) system.
Reads git log since last tag, drafts CHANGELOG entry. Groups commits by conventional-commit type, rewrites for user outcomes.
Auto-generates changelogs from conventional commits. Parses git log since last tag, groups by type, writes user-outcome-focused CHANGELOG.md entries.
© 2026 Flowy · Free and open source
Built for Claude Code · Not affiliated with Anthropic