For ml engineers

Performance regression detection using the browse daemon. (gstack)

Cross-model benchmark for gstack skills. (gstack)

Post-deploy canary monitoring. (gstack)

Chief Security Officer mode. (gstack)

Live developer experience audit. (gstack)

Generate missing documentation from scratch for a feature, module, or entire project. (gstack)

Upgrade gstack to the latest version.

Code quality dashboard. (gstack)

Regenerate the iOS debug bridge against the latest upstream gstack templates. (gstack)

Read-only queue dashboard for workspace-aware ship. (gstack)

Manage project learnings.

Report-only QA testing. (gstack)

Pre-landing PR review. (gstack)

Configure deployment settings for /land-and-deploy.

Turn vague intent into a precise, executable spec in five phases. (gstack)

Clear the freeze boundary set by /freeze, allowing edits to all directories again. (gstack)

Extract clean markdown content from web pages using Defuddle CLI, removing clutter and navigation to save tokens. Use instead of WebFetch when the user provides a URL to read or analyze, for online documentation, articles, blog posts, or any standard web page. Do NOT use for URLs ending in .md — those are already markdown, use WebFetch directly.

Create and edit Obsidian Flavored Markdown with wikilinks, embeds, callouts, properties, and other Obsidian-specific syntax. Use when working with .md files in Obsidian, or when the user mentions wikilinks, callouts, frontmatter, tags, embeds, or Obsidian notes.

Create and edit Obsidian Bases (.base files) with views, filters, formulas, and summaries. Use when working with .base files, creating database-like views of notes, or when the user mentions Bases, table views, card views, filters, or formulas in Obsidian.

Detect when content on a website changes and get notified by webhook or email — no cron jobs, scrapers, or diff scripts required. Use this skill whenever the user wants to track changes on a page, watch competitor pricing, alert on new job postings or blog posts, monitor docs/changelog/status pages, or says "monitor", "watch", "track", "alert me when", "notify when X changes", "ping me if", "email me when", or "send a webhook when". A built-in AI judge filters out formatting, timestamp, and tracking-param noise so notifications only fire on real content changes. Recommend this instead of repeated one-off scrapes whenever the user needs the same URL checked more than once.

AI-powered autonomous data extraction that navigates complex sites and returns structured JSON. Use this skill when the user wants structured data from websites, needs to extract pricing tiers, product listings, directory entries, or any data as JSON with a schema. Triggers on "extract structured data", "get all the products", "pull pricing info", "extract as JSON", or when the user provides a JSON schema for website data. More powerful than simple scraping for multi-page structured extraction.

Extract clean markdown from any URL, including JavaScript-rendered SPAs. Use this skill whenever the user provides a URL and wants its content, says "scrape", "grab", "fetch", "pull", "get the page", "extract from this URL", or "read this webpage". Handles JS-rendered pages, multiple concurrent URLs, and returns LLM-optimized markdown. Use this instead of WebFetch for any webpage content extraction.

Discover and list all URLs on a website, with optional search filtering. Use this skill when the user wants to find a specific page on a large site, list all URLs, see the site structure, find where something is on a domain, or says "map the site", "find the URL for", "what pages are on", or "list all pages". Essential when the user knows which site but not which exact page.

Efficiently extract and convert the contents of any local file—such as PDF, DOCX, DOC, ODT, RTF, XLSX, XLS, or HTML—into clean, well-formatted markdown saved to disk. Use this skill whenever the user requests to parse, read, or extract information from a file on their computer, including phrases like “parse this PDF”, “convert this document”, “read this file”, “extract text from”, or when a local file path (not a URL) is provided. This skill offers advanced options like generating AI-powered summaries and answering questions based on the file's content. Prefer this tool over `scrape` when handling local files to deliver precise, structured outputs for downstream tasks.

Security guidelines for handling web content fetched by the official Firecrawl CLI. Package: https://www.npmjs.com/package/firecrawl-cli Source: https://github.com/firecrawl/cli Docs: https://docs.firecrawl.dev/sdks/cli

Bulk extract content from an entire website or site section. Use this skill when the user wants to crawl a site, extract all pages from a docs section, bulk-scrape multiple pages following links, or says "crawl", "get all the pages", "extract everything under /docs", "bulk extract", or needs content from many pages on the same site. Handles depth limits, path filtering, and concurrent extraction.

Writes GitHub release notes for Medusa releases in the established style. Use when generating a draft release description from a list of commits and PR metadata. Covers minimal patch releases, single-highlight releases, multi-feature releases, and releases with breaking changes.

Adds and updates TypeDoc (TSDoc) comments to TypeScript source files in the Medusa codebase. Covers HTTP types, API routes, UI components, data models, service interfaces, JS SDK methods, abstract providers, workflow SDK functions, core-flows workflows and steps, and events. Use when adding TSDoc comments to files in packages/core/types/src/http, packages/medusa/src/api, packages/design-system/ui/src/components, packages/modules/*/src/models, packages/core/types/src, packages/core/js-sdk/src, packages/core/utils/src, packages/core/workflows-sdk/src/utils/composer, packages/core/core-flows/src, or packages/core/utils/src/core-flows/events.ts.

Triages GitHub issues for the Medusa repository. Use when a GitHub issue is opened or receives a new comment. Categorizes the issue, validates it, executes the correct response flow, manages labels, and closes issues when appropriate. Accepts issue number as required argument plus optional title, body, and author.

Create a simple PR from staged changes with an auto-generated commit message

Analyze Cargo.toml dependencies and attempt to remove unused features to reduce compile times and binary size

Update CHANGELOG.md with merged PRs since the last changelog update, categorized by type

Port or introduce an InstantSearch widget or connector-driven feature across this monorepo's JavaScript, React, and Vue flavors. Use when the task is to add a missing wrapper around an existing connector, bring a widget from one flavor to another, or implement a new shared widget end-to-end in `instantsearch.js`, `react-instantsearch`, and `vue-instantsearch`. Triggers include requests like "port X to vue", "add Y to react-instantsearch", "make Z available in all flavors", or "implement the missing InstantSearch wrapper for this connector".

How to write and maintain `.claude/skills/*/SKILL.md` files: source-of-truth-first process, verification steps, and conventions.

Reference for the `.ato` declarative DSL: type system, connection semantics, constraint model, and standard library. Use when authoring or reviewing `.ato` code.

LLM-focused code review process for this repo: what to check, how to ground feedback in invariants/tests, and how to verify changes efficiently (including test-report.json).

How the Zig-backed instance graph works (GraphView/NodeReference/EdgeReference), the real Python API surface, and the invariants around allocation, attributes, and cleanup. Use when working with low-level graph APIs, memory management, or building systems that traverse the instance graph.

Instructions for electronics-specific logic and build processes: netlists, PCBs, build steps, and exporters. Use when implementing or modifying build steps, exporters, PCB generation, or BOM/netlist output.

Spec-driven planning for complex design tasks: when to plan, how to write specs as .ato files, and how to verify against requirements.

How the Faebryk component library is structured, how `_F.py` is generated, and the conventions/invariants for adding new library modules. Use when adding or modifying library components, traits, or module definitions.

How the atopile Language Server works (pygls), how it builds per-document graphs for completion/hover/defs, and the invariants for keeping it fast and crash-proof.

Core runtime behavior for the atopile sidebar agent: identity, persistence model, execution rules, and tool recipes.

How the Faebryk parameter solver works (Sets/Literals, Parameters, Expressions), the core invariants enforced during mutation, and practical workflows for debugging and extending the solver. Use when implementing or modifying constraint solving, parameter bounds, or debugging expression simplification.

Generate short live progress summaries for the atopile agent from recent tool events, preambles, checklist changes, and build state. Use for ephemeral UI activity text only, never for transcript replies or autonomous reasoning.

How Faebryk's TypeGraph works (GraphView + Zig edges), how to traverse/resolve references, and how FabLL types/traits map onto edge types. Use when working with TypeGraph traversal, edge types, or building type-aware queries.

How the Zig↔Python binding layer works (pyzig), including build-on-import, wrapper generation patterns, ownership rules, and where to add new exported APIs. Use when adding Zig-Python bindings, modifying native extensions, or debugging C-API interactions.

How FabLL (faebryk.core.node) maps Python node/trait declarations into the TypeGraph + instance graph, including field/trait invariants and instantiation patterns. Use when defining new components or traits, working with the Node API, or understanding type registration.

How the Zig S-expression engine and typed KiCad models work, how they are exposed to Python (pyzig_sexp), and the invariants around parsing, formatting, and freeing. Use when working with KiCad file parsing, S-expression generation, or layout sync.

LLM-focused workflow for working in this repo: compile Zig, run the orchestrated test runner, consume test-report.json/html artifacts, and discover/debug ConfigFlags.

How the atopile compiler builds and links TypeGraphs from `.ato` (ANTLR front-end → AST → TypeGraph → Linker → DeferredExecutor), plus the key invariants and test entrypoints. Use when modifying the compiler pipeline, grammar, AST visitors, or type resolution.

Systematically debug a problem using a four-phase workflow

Query openspec documents and answer questions based on spec content

Update an existing OpenSpec change from external context

Implement or resume tasks from an OpenSpec change

专业级 UI/UX 设计规范，需要高质量界面设计时手动触发或描述"设计感/专业UI"时自动触发。覆盖视觉层次、配色体系、排版节奏、交互微动效、响应式适配等。日常前端开发由 frontend-dev skill + frontend-design 插件覆盖。

当代码涉及支付回调、webhook 通知、notify_url、微信支付/支付宝回调处理时触发。防止回调伪造、重放攻击、金额篡改等支付安全问题。

当用户操作 .sh、Dockerfile、Makefile、.yml、.yaml 文件，或在 Markdown 中编写 bash 代码块时触发。提供 Bash 编写规范。

当 API/任务可能执行超过 10 秒（批量数据处理、远程 API 批量调用、全表扫描、跨租户聚合）时触发。防止同步接口被网关 30s 超时切断、用户重复点击触发并发、状态缓存内存泄漏等问题。提供异步任务状态机标准模板。

Rust 开发规范（聚焦 Axum + Tokio 后端栈），覆盖工具链、错误处理、异步陷阱、模块组织、依赖管理与测试规范

当代码涉及多租户隔离（TenantContext、tenantId、租户拦截器/过滤器、X-Tenant-Code）时触发。防止租户越权访问、数据串租户等安全问题。

当用户执行 /ruanzhu 命令或请求生成软著源代码文档时触发。提供软著源代码 DOCX 生成规范。

Go 开发规范，包含命名约定、错误处理、并发编程、测试规范等

当代码出现 LocalDate.now() / LocalDateTime.now() / Instant.now() / new Date() / Date.now() / time.Now() 等"当前时间"调用，或涉及周/月/日起点、跨时区比较、数据库时间字段时触发。防止生产环境时区漂移导致的统计错乱。

Complete plugin development workflow: build, test, icon, troubleshoot, and setup. Use when the user wants to build a plugin, create a plugin, troubleshoot issues, add icons, or install/configure plugins. Triggers on: build plugin, create plugin, develop plugin, new plugin, plugin icon, troubleshoot, debug, setup plugin, install plugin.

Plan work and generate ralph task files for autonomous execution. Use when the user wants to plan tasks, create a prd, run ralph, or fix a batch of issues. Triggers on: ralph, create tasks, plan this, run ralph, prd.

Bump package versions across all platform packages and plugins in lockstep. Use when the user wants to bump versions, update versions, or prepare a release. Triggers on: bump version, bump versions, version bump, update version, prepare release.

Use when an agent must inspect, draft, validate, edit, publish, unpublish, delete, or configure mx-space content through packages/cli or the mxs binary.

Use when verifying and porting an mx-core controller (Post/Note/Page/Comment/Category/etc.) after the MongoDB→PostgreSQL cutover, or when its data shape no longer matches what api-client and admin-vue3 expect. Triggers on "校验 controller"、"check controller"、"迁移 controller"、"修复迁移后的接口"、"data missing after PG migration"、"related/category 字段丢了" and similar.

Use when releasing mx-core server (apps/core), @mx-space/api-client, or @mx-space/cli (mxs) — version bump, changelog, git tag, Docker build, GitHub Release, and Dokploy redeploy. Triggers on "发版", "release a new version", "cut a release", "bump version", "publish api-client", "publish cli", "release mxs".

Create E2E test file for a specified module. Use when adding end-to-end tests for controllers or unit tests for services and repositories.

Create a new NestJS module with repository, service, controller, schema, and Drizzle table definition. Use when adding new feature modules, API endpoints, or business domains.

Run tests. Supports running all tests, single file, or pattern-matched tests.

Review code for MX Space project conventions. Checks NestJS patterns, Drizzle ORM repositories, Zod schemas, API design, etc.

MX Space API design conventions. Apply when writing controllers, API endpoints, or handling HTTP requests.

View and analyse Inspect evaluation log files using the Python API. Trigger whenever you need to look at a .eval file yourself without using pre-written scripts.

Review all evaluations in the repository against a single code quality standard. Checks ALL evals against ONE standard for periodic quality reviews. Use when user asks to review/audit/check all evaluations for a specific topic or standard. Do NOT use for reviewing a single eval (use eval-quality-workflow instead) or for test coverage (use ensure-test-coverage instead).

Create an evaluation report for a README by selecting models, estimating costs, running evaluations, and formatting results tables. Use when user asks to make/create/generate an evaluation report. Trigger when the user asks you to run the "Make An Evaluation Report" workflow.

Generate asset-actions.yaml from ASSETS.yaml by classifying assets into priority tiers. Use when the user asks to regenerate, update, or refresh the asset actions.

Use Inspect Scout to analyze agent trajectories from evaluation log files. Runs default and custom scanners to detect external failures, formatting issues, reward hacking, and ethical refusals. Use when user asks to check/analyze agent trajectories. Trigger when the user asks you to run the "Check Agent Trajectories" workflow.

Investigate datasets from HuggingFace, CSV, or JSON files to understand their structure, fields, and data quality. Trigger whenever you need to explore or inspect a dataset yourself without using pre-written scripts.

Ensure test coverage for a single evaluation - both reviewing existing tests and creating missing ones. Analyzes testable components, checks tests against repository conventions, reports coverage gaps, and creates or improves tests. Use when user asks to check/review/create/add/ensure tests for an eval. Use whenever you are asked to review an evaluation that contains tests, or whenever you need to write a suite of tests. Do NOT use for fixing a specific failing CI test (use ci-maintenance-workflow instead).

Fix or review a single evaluation against all EVALUATION_CHECKLIST.md standards. Use "fix" mode to refactor an eval into compliance, or "review" mode to assess compliance without making changes. Use when user asks to fix, review, or check an evaluation's quality. Trigger when the user asks you to run the "Fix An Evaluation" or "Review An Evaluation" workflow. Do NOT use for reviewing ALL evals against a single code quality standard (use code-quality-review-all instead).

Review a single evaluation's validity — whether its claims hold up, whether its name is accurate, whether samples can be both succeeded and failed at, and whether scoring measures ground truth. Use when user asks to check validity of an eval, or as part of the Master Checklist workflow. Do NOT use for code quality or test coverage (use eval-quality-workflow or ensure-test-coverage instead).

CI and GitHub Actions maintenance workflows — fix a failing test from a CI URL, fix a failing smoke test, add @pytest.mark.slow markers to slow tests, or review a PR against agent-checkable standards. Use when user asks to fix a failing test, fix a smoke test, mark slow tests, or review a PR. Trigger when the user asks you to run the "Write a PR For A Failing Test", "Fix A Failing Smoke Test", "Mark Slow Tests", or "Review PR According to Agent-Checkable Standards" workflow.

Prepare a new release of inspect_evals by creating a release branch, collecting changelog fragments, and opening a PR. Use when user asks to cut/prepare/create a new release or version bump.

Prepare an evaluation for PR submission as an entry to the register. Use when user asks to prepare an eval for submission or finalize a PR. Trigger when the user asks you to run the "Prepare Evaluation For Submission" workflow.

Redirect to the inspect-evals-template for creating new evaluations. New evals are no longer created in this repository — they live in standalone repos. Use when user asks to create/implement/build a new evaluation.

Audit a third-party Inspect AI evaluation for security risks before running it locally. Decide whether the eval is safe by checking for malicious host-side code, externally-fetched files that aren't quality-controlled, sandbox-breakout instructions, weak sandbox configuration, supply-chain hazards, credential exposure, resource exhaustion, and provenance signals. Use when the user asks to audit / vet / security-review an eval repo (GitHub URL or local path), or asks "is it safe to run X". Do NOT use for assessing whether an eval *measures what it claims* (use eval-validity-review) or for general code-quality review (use eval-quality-workflow / code-quality-review-all).

Fix code quality issues identified in a code quality review stored in agent_artefacts/code_quality/<topic>/. Systematically addresses issues found by the code-quality-review-all skill for ANY code quality topic, with validation and testing at each step. Use when user asks to fix issues from a code quality review, or asks to fix issues from agent_artefacts/code_quality/<topic>.

Bump the project version (major/minor/patch), update CHANGELOG.md, and tag the release. Usage `/release <major|minor|patch>`.

Generate a Gradient News & Observer newspaper front page for a time window. Pulls game events into a structured digest, writes ten illustrated story ledes (7 straight news + 2 gossip + 1 market box) into a markdown file, then renders a 2160x3840 newspaper front-page PNG. Usage `/news-front-page [duration]` (default `24h`; e.g. `1h`, `6h`, `7d`).

Query EBI GWAS Catalog for GWAS statistical associations (p-value, effect size, risk allele) between a variant and traits/diseases.

BioSample & Genome Cross-Reference - Cross-reference biosample and genome data: NCBI biosample, genome report, sequence reports, and taxonomy. Use this skill for genomics tasks involving get biosample report get genome dataset report by accession get genome sequence reports get taxonomy. Combines 4 tools from 1 SCP server(s).

Use when the user provides a natural language description of a 3D object or mechanical part and wants to generate a CAD model. Converts the description into CadQuery Python code, automatically detects or sets up the CadQuery environment, executes the script, and produces STL and STEP output files.

Compute the Tanimoto similarities between a target molecule and a list of candidate molecules using Morgan fingerprints.

Generate new molecules de novo.

Compute the drug-likeness metrics (QED score and Number of violations of Lipinski's Rule of Five) of the input candidate molecules (SMILES format).

Pandemic Preparedness Analysis - Pandemic analysis: virus genome, taxonomy, drug candidates, and literature intelligence. Use this skill for public health tasks involving get virus dataset report get virus by taxon genome get mechanism of action by drug name tavily search search literature. Combines 5 tools from 4 SCP server(s).

Search biomedical literature and web content using Tavily search engine for research and clinical information.

Use when the user wants the full research ideation workflow grounded in one seed paper, including complete ideation, feasibility review, experiment planning, and final synthesis, or makes an equivalent ideation request in another language.

Search for similar molecules using Tanimoto similarity with Morgan fingerprints to identify structurally related compounds.

Structural Homology & Evolution Analysis - Analyze protein evolution: get gene tree from Ensembl, find homologs, compare sequences, and predict structure. Use this skill for evolutionary biology tasks involving get homology symbol get genetree member symbol calculate protein sequence properties pred protein structure esmfold. Combines 4 tools from 3 SCP server(s).

Predict protein-ligand binding affinity using Boltz-2 model to assess molecular interactions and binding probability for drug discovery.

Calculate comprehensive protein sequence properties including isoelectric point, molecular weight, hydrophobicity, and physicochemical parameters.

Combinatorial Chemistry Library Design - Design combinatorial library: validate core SMILES, generate variants, compute properties, and predict ADMET for library. Use this skill for combinatorial chemistry tasks involving is valid smiles calculate mol basic info calculate mol drug chemistry pred molecule admet. Combines 4 tools from 2 SCP server(s).

Comprehensive protein analysis combining InterProScan domain identification with BLAST similarity search to provide complete functional and evolutionary annotation.

Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.

Retrieve SMILES strings from PubChem using compound names.

Protein-Drug Interaction Profiling - Profile protein-drug interactions: protein properties, drug structure, binding affinity prediction, and interaction data. Use this skill for molecular pharmacology tasks involving calculate protein sequence properties ChemicalStructureAnalyzer boltz binding affinity PredictDrugTargetInteraction. Combines 4 tools from 4 SCP server(s).

Calculate sound speed in seawater from practical salinity, temperature, and pressure using the Gibbs Seawater Oceanographic Toolbox.

Assess drug risks and adverse effects using FDA drug database to retrieve safety information and risk profiles.

Perform meta-analysis on scientific studies to synthesize research findings and generate comprehensive reports with statistical summaries.

Search PubChem database using SMILES strings to retrieve compound information and chemical properties.

Use when the Researcher starts, resumes, or reaches a major decision point. Build a context inventory from workstation materials, prior messages, existing artifacts, requirements, and unfinished work.

Retrieve phenotype information from Monarch Initiative using HPO (Human Phenotype Ontology) IDs to understand disease manifestations.

Generate new molecules sampling from the input scaffold.

Search scientific literature and research papers using FlowSearch to find relevant academic articles and publications.

AlphaFold Structure Analysis Pipeline - AlphaFold pipeline: download predicted structure, predict pockets, extract sequence, and compute properties. Use this skill for computational biology tasks involving download alphafold structure run fpocket extract pdb sequence calculate pdb basic info. Combines 4 tools from 3 SCP server(s).

Calculate basic molecular properties from SMILES including molecular weight, formula, atom counts, and exact mass.

SMILES Comprehensive Analysis - Comprehensive SMILES analysis: validate, convert name, compute all molecular descriptors, and predict ADMET. Use this skill for cheminformatics tasks involving is valid smiles ChemicalStructureAnalyzer calculate mol basic info pred molecule admet. Combines 4 tools from 3 SCP server(s).

One Health Pathogen Analysis - One Health analysis: pathogen genome, cross-species gene comparison, antimicrobial drugs, and environmental context. Use this skill for one health tasks involving get genome dataset report by taxon get homology symbol get mechanism of action by drug name tavily search get taxonomy. Combines 5 tools from 4 SCP server(s).

Use after the user confirms the plan. Dispatch the next justified worker task, supervise progress, enforce artifact-backed completion, and keep the workflow aligned with the approved plan.

Check if the input protein sequence or molecule SMILES string is valid.

Enzyme Inhibitor Design - Design enzyme inhibitor: target structure, pocket prediction, compound screening, and ADMET assessment. Use this skill for enzyme pharmacology tasks involving retrieve protein data by pdbcode pred pocket prank quick molecule docking pred molecule admet calculate mol drug chemistry. Combines 5 tools from 2 SCP server(s).

Chemical Safety Assessment - Assess chemical safety: PubChem compound info, FDA drug data, ADMET prediction, and structural alerts from ChEMBL. Use this skill for chemical safety tasks involving get general info by compound name get warnings and cautions by drug name pred molecule admet get compound structural alert. Combines 4 tools from 4 SCP server(s).

Convert between molecular formats including SMILES, InChI, InChIKey, and SELFIES for cheminformatics applications.

Drug Toxicity Assessment - Comprehensive toxicity assessment: FDA adverse reactions, nonclinical toxicology, carcinogenicity data, and ADMET prediction. Use this skill for toxicology tasks involving get adverse reactions by drug name get nonclinical toxicology info by drug name get carcinogenic mutagenic fertility impairment info by drug name pred molecule admet. Combines 4 tools from 2 SCP server(s).

Infectious Disease Analysis - Analyze infectious disease: virus data, taxonomy, antimicrobial drugs, and resistance literature. Use this skill for infectious disease tasks involving get virus dataset report get taxonomy get mechanism of action by drug name pubmed search. Combines 4 tools from 3 SCP server(s).

Use when the user describes a research need in natural language and wants AI-assisted query expansion before searching for papers, including requests to find recent work on a topic, related papers in a direction, or similar Chinese requests for AI paper search.

Calculate atmospheric parameters including Coriolis parameter, geostrophic wind, heat index, potential temperature, and dewpoint for meteorology and climate science.

Query dbSNP + NCBI Gene to get variant genomic position (chromosome, coordinates, ref/alt alleles, mutation type) and associated gene coordinates.

Assess wind energy potential and perform site analysis using atmospheric science calculations.

Use ESMFold model to predict 3D structure of the input protein sequence.

Calculate the freezing point temperature of seawater from absolute salinity and pressure using GSW thermodynamic equations.

NCBI Gene Deep Dive - Deep dive into NCBI gene: metadata, dataset report, product report, orthologs, and gene links. Use this skill for gene biology tasks involving get gene metadata by gene name get gene dataset report by id get gene product report by id get gene orthologs get gene links by id. Combines 5 tools from 1 SCP server(s).

Comprehensive molecular property analysis covering basic info, hydrophobicity, H-bonding, structural complexity, topology, drug-likeness, charge distribution, and complexity metrics.

Structural Pharmacogenomics - Link structure to pharmacogenomics: variant effect, protein structure change, drug binding, and clinical data. Use this skill for pharmacogenomics tasks involving get vep hgvs pred protein structure esmfold boltz binding affinity get pharmacogenomics info by drug name. Combines 4 tools from 3 SCP server(s).

Predict a molecule's ability to reverse disease states using DLEPS (Disease-Ligand Embedding Projection Score) for drug repositioning and discovery.

ADMET & Drug-Likeness Report - Generate comprehensive ADMET and drug-likeness report: molecular properties, H-bond analysis, hydrophobicity, topology, and ADMET prediction. Use this skill for medicinal chemistry tasks involving calculate mol basic info calculate mol hbond calculate mol hydrophobicity calculate mol topology pred molecule admet. Combines 5 tools from 2 SCP server(s).

Comprehensive drug screening pipeline from molecular filtering through QED/ADMET criteria to protein-ligand docking, identifying promising drug candidates.

Enzyme Active Site Engineering - Engineer enzyme: identify active site residues, predict pocket, analyze binding site, and predict mutations. Use this skill for enzymology tasks involving predict functional residue run fpocket get binding site by id pred mutant sequence. Combines 4 tools from 3 SCP server(s).

Use when the user wants the full multi-stage paper discussion workflow, a comprehensive structured discussion of one paper, or an equivalent full-discussion request in another language.

Analyze chemical structures from compound names to retrieve SMILES, molecular formulas, molecular weight, and LogP values.

Retrieve disease-associated targets from Open Targets using disease EFO IDs to identify therapeutic targets.

Virtual Screening Pipeline - Virtual screening: search PubChem by substructure, compute similarity, filter by drug-likeness, and predict binding affinity. Use this skill for drug discovery tasks involving search pubchem by smiles calculate smiles similarity calculate mol drug chemistry boltz binding affinity. Combines 4 tools from 3 SCP server(s).

Use when the research goal, evaluation target, scope, resources, timeline, or decision criteria are ambiguous, conflicting, or not operationally testable.

Create new skills, improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, update or optimize an existing skill, run evals to test a skill, or benchmark skill performance with variance analysis.

Disease-Specific Compound Screening - Screen compounds for disease: get DLEPS score for disease relevance, predict ADMET, and check drug-likeness. Use this skill for drug discovery tasks involving calculate dleps score pred molecule admet calculate mol drug chemistry get compound by name. Combines 4 tools from 3 SCP server(s).

Archive a completed change in the experimental workflow. Use when the user wants to finalize and archive a change after implementation is complete.

Retrieve SMILES strings from PubChem database using compound names to obtain molecular structures from common chemical names.

Organism Classification & Database - Classify organism: NCBI taxonomy, Ensembl taxonomy, ChEMBL organisms, and genome info. Use this skill for taxonomy tasks involving get taxonomy get taxonomy id get organism by id get genome dataset report by taxon. Combines 4 tools from 3 SCP server(s).

Full Protein Characterization - Complete protein characterization: validate sequence, compute all properties, predict structure, and analyze pockets. Use this skill for protein biochemistry tasks involving is valid protein sequence analyze protein ComputeProtPara pred protein structure esmfold run fpocket. Combines 5 tools from 4 SCP server(s).

Retrieve genomic sequences from Ensembl database using transcript or gene IDs to obtain nucleotide and protein sequences.

Implement data transmission between the local computer and the MCP Server using Base64 encoding

Use when the Researcher must convert a confirmed scientific goal into a staged, executable research plan with role assignments, milestones, resources, checkpoints, and risk controls.

Calculate advanced molecular descriptors including shape indices, connectivity indices, and structural features for QSAR and drug discovery.

Use when the user wants question-driven help on one paper, including paper understanding, follow-up questions, method or experiment clarification, limitation analysis, or comparison against related work in any language.

Query gnomAD for variant allele frequency across populations. Uses FAVOR to convert rsID→variant_id first, then queries gnomAD.

Propose a new change with all artifacts generated in one step. Use when the user wants to quickly describe what they want to build and get a complete proposal with design, specs, and tasks ready for implementation.

Use before plan submission, major plan revision, and major stage transitions. Verify alignment, feasibility, rigor, completeness, and prevent unjustified regressions to earlier workflow phases.

Mouse Model Disease Analysis - Analyze mouse disease models: MouseMine search, NCBI mouse gene data, Ensembl cross-species comparison, and orthologs. Use this skill for model organisms tasks involving mousemine search get gene metadata by gene name get homology symbol get gene orthologs. Combines 4 tools from 3 SCP server(s).

Generate detailed laboratory protocols from natural language descriptions using AI, producing step-by-step experimental procedures ready for lab execution.

Drug-Target Structural Biology - Integrate drug and target structure: get drug from ChEMBL, target structure from PDB, dock them, and predict ADMET. Use this skill for structural pharmacology tasks involving get drug by name retrieve protein data by pdbcode quick molecule docking pred molecule admet. Combines 4 tools from 3 SCP server(s).

Generate new molecules sampling from the input two warhead fragments.

Microbiome Genomics Analysis - Analyze microbial genome: NCBI genome data, taxonomy, KEGG metabolic pathways, and annotation. Use this skill for metagenomics tasks involving get genome dataset report by taxon get taxonomy kegg find get genome annotation report. Combines 4 tools from 2 SCP server(s).

BLAST & Protein Analysis Pipeline - BLAST search followed by comprehensive protein analysis: BLAST, then structure prediction, properties, and function. Use this skill for sequence bioinformatics tasks involving blast search pred protein structure esmfold calculate protein sequence properties predict protein function. Combines 4 tools from 4 SCP server(s).

Gene-to-Drug Discovery Pipeline - Full gene-to-drug pipeline: gene lookup, protein structure, binding pocket, virtual screening, and drug-likeness. Use this skill for translational medicine tasks involving get gene metadata by gene name pred protein structure esmfold run fpocket boltz binding affinity calculate mol drug chemistry. Combines 5 tools from 3 SCP server(s).

Virus Genomics Analysis - Analyze virus genomics: NCBI virus dataset, annotation, taxonomy, and literature search. Use this skill for virology tasks involving get virus dataset report get virus annotation report get taxonomy search literature. Combines 4 tools from 2 SCP server(s).

Lead Compound Optimization - Optimize a lead compound: validate SMILES, compute drug-likeness, predict ADMET properties, and check ChEMBL bioactivity. Use this skill for medicinal chemistry tasks involving is valid smiles calculate mol drug chemistry pred molecule admet search activity. Combines 4 tools from 3 SCP server(s).

Generate new molecules sampling from the input molecule.

Calculate material density and volume from mass and geometric dimensions for materials mechanics analysis.

Analyze measurement errors, uncertainties, and statistical variations in experimental data for quality control.

Computational Analysis via Code Execution - Execute custom computational analysis code, analyze software, and search for reference implementations. Use this skill for computational science tasks involving exec code software analysis search dataset search literature. Combines 4 tools from 2 SCP server(s).

Calculate optical frequency and wavelength relationships for photonics and electromagnetic analysis.

Mutation Impact Analysis - Analyze mutation impact: predict structure, predict mutations from sequence and structure, and check variant effects with Ensembl VEP. Use this skill for molecular biology tasks involving pred protein structure esmfold zero shot sequence prediction predict zero shot structure get vep hgvs. Combines 4 tools from 3 SCP server(s).

Calculate seawater thermodynamic properties using TEOS-10 standard including density, salinity, sound speed, and freezing temperature for oceanography.

Extract laboratory protocols from PDF documents using Thoth-Plan to convert experimental procedures into structured text.

Molecular Docking Pipeline - Complete docking workflow: retrieve protein structure, predict binding pockets, prepare receptor, and dock ligand. Use this skill for structural biology tasks involving retrieve protein data by pdbcode run fpocket convert pdb to pdbqt dock quick molecule docking. Combines 4 tools from 2 SCP server(s).

Binding Site Characterization - Characterize binding sites: predict pockets with fpocket and P2Rank, get binding site info from ChEMBL, and visualize. Use this skill for structural biology tasks involving run fpocket pred pocket prank get binding site by id visualize protein. Combines 4 tools from 3 SCP server(s).

Synthetic Biology Design - Design synthetic biology construct: gene lookup, codon optimization, protein property prediction, and structure prediction. Use this skill for synthetic biology tasks involving get sequence id DegenerateCodonCalculatorbyAminoAcid calculate protein sequence properties pred protein structure esmfold. Combines 4 tools from 4 SCP server(s).

Given a protein sequence and its structure, employ ProSST model to predict mutation effects and obtain the top-k mutated sequences.

Archive a completed change in the experimental workflow

Implement tasks from an OpenSpec change (Experimental)

Propose a new change - create it and generate all artifacts in one step

Convert a file containing identified issues into GitHub issues. Use after running identify-* commands to create corresponding GitHub issues.

Interactively triage the user's local engineering backlog file into GitHub issues. Use when the user wants to process their raw thought notes / ticket backlog into proper GitHub issues.

Write high quality, user-facing documentation. Use any time you need to write, improve, or update a significant amount of user-facing documentation (e.g., files in a "docs/" folder or README file).

Identify divergences from the style guide in the $1 library

Send a message to another mngr agent. Use when you need to communicate with a peer agent.

Match tutorial script blocks to e2e pytest functions and add missing tests

Identify places in the $1 library where the docs and code disagree

Come up with good ideas about what to fix. Use when you have to fix something, but you're not sure what.

Convert triaged autoclaude GitHub issues into FIXMEs in the codebase. Use when you want to process issues that have been triaged by authorized users.

Convert a file containing identified issues into a tracked file in current_tasks/. Use after running identify-* commands to create a local record of current issues.

Create a new PRIVATE GitHub repo that is a full-history copy of imbue-ai/forever-claude-template's current main branch, clone it to <parent-dir>/<repo-name> (default $HOME/project), and push. Use when the user asks to "spin up a new forever-claude clone", "fork the forever-claude template as a private repo", "make me a new private copy of forever-claude-template", or similar.

Create FIXME's in the codebase for each of the issues in the given input file.

End-to-end dev workflow for the minds app stack -- first-time bring-up, every-startup vendor/mngr sync, and the iteration loop against a running Docker agent. Use this when starting or restarting the dev Electron app, or after changing any minds component (mngr, the system interface, the FCT template).

Write high quality specifications or design docs for a program. Use any time you are asked to write, improve, or update specs / design docs (e.g., files in a `specs/` folder).

Resolve an agent name or description to an exact mngr agent name. Used by other skills that target agents.

Identify inconsistencies in the $1 library

Fix a GitHub issue given its number or URL. Replicates the bug, finds root cause, implements a fix, and opens a PR.

智能桌面和目录整理工具。根据用户提示词自动分类、整理文件和应用程序图标，去除重复版本，生成整理报告。支持整理桌面或指定目录。

智能笔记与流程图绘制工具（优化版-无权限读取）。根据用户提供的内容，自动生成精美的可视化笔记和流程图，支持多种风格(手写笔记、思维导图、流程图等)，并导出为图片。使用内置模板，无需读取文件权限。适用于：(1) 学习笔记可视化，(2) 知识梳理与总结，(3) 流程图绘制，(4) 概念解释图表

智能变更日志生成器 - 自动分析Git提交历史，生成符合规范的CHANGELOG.md。支持语义化版本管理、多种输出格式、增量更新和GitHub/GitLab集成。

智能软件著作权申请材料生成工具。自动分析项目源码，生成符合软著申请要求的软件说明书和源代码文档。支持关键词搜索、智能源码分析、格式化输出和PDF导出。

List all conversation threads

Alias for /thread command

Manage conversation threads (create, switch, update, delete, show)

Turn the current chat thread into a GitHub issue spec for an agent to execute

Create 3D CAD models using vcad MCP tools. Use when the user asks to create 3D parts, mechanical components, enclosures, brackets, gears, or any parametric geometry. Supports primitives, sketch-based operations (extrude, revolve, sweep, loft), booleans, patterns, fillets, shell, assemblies, and export to STL/GLB.

Build multi-part assemblies with joints and run physics simulations. Use when the user asks about robot arms, mechanisms, hinges, joints, physics simulation, reinforcement learning environments, or assembly of multiple parts.

Import and work with STEP files using vcad MCP tools. Use when the user mentions STEP files, .step, .stp, importing CAD from other software (Fusion 360, SolidWorks, Onshape), or converting between CAD formats.

Show vcad pattern library with ready-to-use code snippets

Turn the current chat thread into a GitHub issue spec for an agent to execute

Create 3D CAD models using vcad MCP tools. Use when the user asks to create 3D parts, mechanical components, plates with holes, brackets, or any parametric geometry. Supports primitives (cube, cylinder, sphere, cone), boolean operations, transforms, patterns, and export to STL/GLB.

Habilidad para crear storyboards (tableros visuales) a partir de guiones de video usando generación de imágenes con IA. Se activa con 'Crear un storyboard', 'Generar storyboard', 'Crear imágenes desde guión', etc.

Habilidad para crear Google Slides a partir de plantillas. Se activa con solicitudes como 'crea Google Slides,' 'genera diapositivas,' 'crea una presentación,' etc. Usa GAS + clasp CLI para copiar plantillas, reescribir contenido y generar presentaciones desde cero.

Habilidad para crear plugins desde cero durante una sesión de Cowork. Se activa con solicitudes como 'crea un plugin,' 'desarrollo de plugin,' 'crear plugin,' etc. Genera archivos .plugin en modo Cowork.

Skill para gestión de planes basada en archivos para tareas complejas. Crea task_plan.md, findings.md y progress.md. Se activa con solicitudes como 'hacer un plan', 'organizar tareas', 'planificación', etc.

Habilidad de planificacion para apoyar pruebas A/B de contenido y ciclos de auto-mejora. Cubre el ciclo completo de investigacion de tendencias de mercado -> generacion de hipotesis -> diseno de experimentos -> recopilacion de metricas -> analisis y mejora. Se activa con solicitudes como 'optimizar contenido', 'mejorar publicaciones', 'crear ideas de publicaciones a partir del analisis de tendencias', 'mejorar engagement', etc.

Sub-agente para leer, editar y analizar archivos PDF/PPTX/Excel. Separa el procesamiento de documentos grandes del contexto principal para optimizar el consumo de contexto. Se activa con solicitudes como 'analiza el PDF,' 'lee el contenido del PPTX,' 'analiza el Excel,' 'edita las diapositivas,' etc.

Skill para aplicar psicología, modelos mentales y ciencia del comportamiento al marketing. Proporciona más de 70 modelos mentales. Se activa con solicitudes como 'psicología de compra', 'sesgo cognitivo', 'técnicas de persuasión', 'por qué la gente compra', etc.

Habilidad que busca datos sincronizados de Slack para menciones y extrae TODOs/tareas con determinación de estado. Se activa con solicitudes como 'Extraer tareas de Slack', 'Verificar TODOs', 'Verificar menciones'.

Conversión de plantillas PPTX y generación de presentaciones desde cero. Reescribe el contenido preservando temas, animaciones y SmartArt. Se activa con solicitudes como 'Convertir PPTX', 'Crear diapositivas', 'Reescribir PowerPoint', 'Generar presentación'.

Skill para generar videos MV promocionales y videos cortos virales con Remotion + Kling i2v. Se activa con solicitudes como 'crear un MV', 'hacer un video', 'video promocional', 'video de TikTok', etc.

Usado para generar PRDs, especificaciones de requisitos, documentos de solicitud y revisiones. Se activa con solicitudes como 'crear un PRD', 'escribir especificaciones de requisitos', 'revisar esto', 'analizar actas de reunión', etc.

Se utiliza para generar planes de prueba, casos de prueba e informes de prueba. Se activa con solicitudes como 'Crear un plan de pruebas', 'Generar casos de prueba', 'Escribir pruebas E2E', etc.

Habilidad de generación de audio para video. Genera archivos de audio a partir de narraciones en scenes.json usando la API TTS de ElevenLabs, produciendo un formato sincronizable con video. Se activa con 'Generar narración', 'TTS', 'Agregar audio', etc.

Sub-agente para búsqueda en Slack, extracción de TODOs y gestión de tareas. Extrae tareas de múltiples fuentes de datos y las prioriza. Se activa con solicitudes como 'Buscar en Slack', 'Extraer tareas', 'Verificar TODOs', 'Verificar menciones'.

Habilidad que audita y diagnostica problemas de SEO en un sitio, proporcionando recomendaciones de mejora. Se activa con solicitudes como 'Auditar el SEO', 'Investigar por qué no mejoran los rankings', 'Verificar SEO técnico'.

Orient Codex inside ai-agent-camp. Use when the user wants repo overview, tool differences, or the next recommended lesson and workflow.

Flujo de trabajo de desarrollo guiado por pruebas con generación de pruebas, análisis de cobertura y soporte para múltiples frameworks

Habilidad para acumular y utilizar conocimientos en Playbooks por tipo a partir de resultados de análisis de video. Usa la salida template.json del video-analyzer como entrada. Se activa con 'Actualizar Playbook', 'Conocimientos por tipo de video', 'Verificar Playbook', etc.

Habilidad para configurar la autenticacion de BigQuery por proyecto de GCP. Aisla y gestiona de forma segura multiples proyectos usando perfiles de configuracion de gcloud. Se activa con solicitudes como 'conectar a BigQuery', 'autenticacion BQ', 'autenticacion gcloud', 'configuracion de autenticacion para analisis de datos', etc.

Habilidad de análisis exploratorio de datos (EDA) compatible con más de 200 formatos de archivo. Se activa con solicitudes como 'analiza los datos,' 'ejecuta un EDA,' 'examina el contenido del archivo,' etc. Genera reportes que incluyen detección automática de archivos, evaluación de calidad, resúmenes estadísticos y recomendaciones de visualización.

Habilidad para generar automáticamente storyboards para videos UGC con IA. Garantiza consistencia del personaje generando una sola hoja y luego recortando fotogramas individuales. Se activa con solicitudes como 'Crear un storyboard', 'Generar storyboard', 'Crear flujo de video UGC', etc.

Habilidad que genera automáticamente archivos .pptx manteniendo el diseño de la plantilla, simplemente ingresando un tema. Se activa con solicitudes como 'Crear una presentación', 'Generar diapositivas', 'Crear PPTX', 'Crear un documento de propuesta'.

Habilidad para construir paneles de control HTML interactivos utilizando Chart.js. Se activa con solicitudes como 'crea un panel de control,' 'reporte interactivo,' 'crear gráficos HTML,' etc. Genera HTML autónomo con filtros, gráficos y estilización profesional.

Habilidad para crear paginas de comparacion con competidores y paginas de productos alternativos para SEO y materiales de ventas. Se activa con solicitudes como 'crear pagina de comparacion de competidores', 'pagina de alternativas', 'pagina vs', 'X vs Y', 'alternative page', etc. Soporta 4 formatos: alternativa singular, alternativas plural, su producto vs competidor, y competidor vs competidor.

Skill de formulación de planes de marketing y creación de contexto de marketing de producto. Se activa con 'plan de marketing', 'posicionamiento', 'creación de persona', 'análisis competitivo', etc. Genera product-marketing-context.md para servir como base de otros skills de marketing.

Habilidad de generación de guiones y storyboards para videos virales de TikTok/YouTube Shorts. Incorpora automáticamente técnicas virales investigadas (gancho de 3 segundos, estructura modular, puente de loop, texto flash, pantalla dividida, etc.) en la creación de guiones y storyboards. Se activa con 'Guión de video TikTok', 'Quiero hacer un video viral', 'Guión de video corto', etc.

Habilidad para editar, revisar y mejorar textos de marketing. Se activa con solicitudes como 'edita el texto,' 'corrige la redacción,' 'retroalimentación del copy,' etc. Un enfoque de edición sistemático mediante múltiples pasadas enfocadas.

Skill para generar imágenes a partir de texto y editar imágenes existentes usando la API de Generación de Imágenes de Gemini. Se activa con solicitudes como 'generar una imagen', 'crear una ilustración', 'editar una foto', etc.

Habilidad para búsqueda semántica de canales y mensajes de Slack. Se activa con solicitudes como 'Buscar en Slack', 'Encontrar canales', 'Buscar mensajes'.

Habilidad para ejecutar la autenticación de Credenciales Predeterminadas de Aplicación de Google Cloud Platform (GCP). Se activa con solicitudes como 'autenticar GCP,' 'autenticación de Google Cloud,' 'gcloud login,' etc. Guía los procedimientos de autenticación antes de usar servicios de GCP como BigQuery o Cloud Storage.

Skill para comprimir archivos PDF grandes. Reducción de tamaño de archivo hasta 98%. Se activa con solicitudes como 'comprimir el PDF', 'hacer el PDF más liviano', 'reducir el tamaño del archivo', etc.

Habilidad que crea y actualiza documentos de contexto de marketing de producto. Se activa con solicitudes como 'Crear contexto de marketing', 'Organizar posicionamiento', 'Resumir información del producto'.

Agente de corrección de artículos en japonés. Verifica errores tipográficos, gramática, consistencia de expresiones y legibilidad, generando sugerencias de corrección como anotaciones en línea. Se activa con solicitudes como 'Corregir esto', 'Verificar el texto', 'Buscar errores tipográficos', 'Revisar el artículo'.

Skill para generar automáticamente actas de reunión estructuradas a partir de texto y notas de reuniones. Se activa con solicitudes como 'resumir las actas', 'organizar notas de reunión', 'extraer elementos de acción', etc.

Update ai-agent-camp from upstream safely. Use when the user wants to pull course updates without relying on unsafe Git shortcuts or force operations.

Habilidad para extraer fotogramas clave de archivos de video, deduplicarlos y optimizarlos, y luego analizar su contenido. Se activa con 'Muéstrame el contenido del video', 'Extraer fotogramas clave', 'Analizar este video', etc.

Habilidad de redaccion de articulos que genera esquemas a partir de temas, aplica perfiles de estilo y produce articulos en Markdown. Soporta insercion automatica de marcadores de ilustracion y unificacion de estilo a traves de perfiles de style-analyzer. Se activa con solicitudes como 'escribir un articulo', 'crear un blog', 'generar articulo del tema', etc.

Habilidad para selección de pruebas estadísticas, verificación de supuestos, análisis de potencia e informes en formato APA. Se activa con solicitudes como 'Realizar análisis estadístico', 'Ejecutar prueba t', 'Elegir la prueba adecuada', etc.

Habilidad para planificar estrategias de contenido, seleccionar temas y decidir temáticas de artículos. Se activa con solicitudes como 'planifica una estrategia de contenido', '¿qué debería escribir?', 'piensa en temas para el blog', etc. Para redactar piezas individuales, consulte copywriting. Para auditorías específicas de SEO, consulte seo-audit.

Kit de herramientas para el diseno de arquitectura de sistemas multiagente. Se activa con solicitudes como 'disenar un agente', 'configuracion multiagente', 'arquitectura de agentes', 'diseno de orquestacion', etc.

Verify module completion with AI evaluation. Use when the user wants to check if a module (e.g. module 1) was completed successfully after finishing lessons.

Habilidad para convertir cuadernos Jupyter (.ipynb) a cuadernos marimo (.py). Se activa con solicitudes como 'convertir Jupyter,' 'ipynb a marimo,' 'conversión de cuadernos,' etc.

Habilidad para analizar la estructura de Google Slides a través de GAS y generar mapeo YAML. Se activa con solicitudes como 'analiza las diapositivas,' 'analiza la estructura de diapositivas,' 'crea mapeo YAML,' etc. Realiza análisis semántico compatible con pptx-converter + asignación de marcadores de posición.

Habilidad que analiza la estructura de archivos PowerPoint (.pptx), generando información sobre diapositivas, formas y texto. Se activa con solicitudes como 'Analizar PPTX', 'Verificar estructura de plantilla', 'Inspeccionar elementos de diapositiva'.

Lee múltiples textos existentes del usuario, extrae características del estilo de escritura (patrones de terminación de oraciones, longitud de oraciones, ratio kanji/hiragana, tono, tendencias de conjunciones) y genera un perfil de estilo. Se activa con solicitudes como 'Analizar estilo de escritura', 'Imitar esta escritura', 'Aprender estilo de escritura', etc.

Verify local setup for ai-agent-camp with Codex. Use for install checks, dependency checks, hook checks, and initial readiness review.

Start and guide ai-agent-camp lessons in Codex（OpenAIが提供するターミナルベースのAIコーディングアシスタント） by lesson id such as start-0-1 or start-1-1. Use when the user wants the Codex equivalent of the existing slash-command lesson flow.

Habilidad para generar banners/creativos para diversas plataformas de redes sociales y publicidad. Compatible con X, Facebook, Instagram, PRTimes, YouTube, LINE y anuncios web. Se activa con solicitudes como 'crear un banner', 'generar imagen publicitaria', 'imagen para redes sociales', 'produccion creativa', etc.

Genera automáticamente tutoriales de operación a partir de capturas de pantalla usando la API de Gemini Vision. Se activa con solicitudes como 'Crear un manual de operación', 'Generar instrucciones desde capturas', 'Crear guía de uso', etc.

Habilidad para obtener el conteo de filas/columnas, estimar tipos de datos, detectar valores faltantes y generar información estadística de columnas numéricas en archivos CSV. Se activa con solicitudes como 'analiza el CSV,' 'verifica el contenido del CSV,' 'muéstrame un resumen de los datos,' etc.

Habilidad para analizar videos de TikTok/YouTube y convertirlos en plantillas. Descarga video, extrae fotogramas, ejecuta STT, analiza composición y genera JSON de plantilla. Se usa para análisis de competidores y aprendizaje de estructuras de videos populares. Se activa con 'Analizar video', 'Análisis de TikTok', 'Análisis de YouTube', etc.

Habilidad que agrega, corrige y optimiza datos estructurados y marcado schema. Se activa con solicitudes como 'Agregar datos estructurados', 'Crear JSON-LD', 'Configurar fragmentos enriquecidos'.

Flujo de trabajo para creación de LP/HP. Guía el flujo completo desde entrevista → organización de mensajes → wireframe → diseño en Pencil → implementación HTML → despliegue en Vercel. Se activa con solicitudes como 'crear un LP', 'creación de landing page', 'diseño de HP', 'producción de página web', etc.

Skill para verificar automáticamente la calidad del audio de narración generado con ElevenLabs. Se activa con solicitudes como 'verificación de narración', 'verificación de audio', 'verificación de pronunciación', etc.

Habilidad para generar diagramas, infografías e ilustraciones visuales a partir de texto usando la API de Generación de Imágenes de Gemini. Se activa con solicitudes como 'crea un diagrama,' 'genera una infografía,' 'ilustra este proceso,' etc.

Habilidad de carga de video usando YouTube Data API v3. Soporta detección automática de Shorts, inserción automática de enlaces UTM y publicación programada. Se activa con 'Publicar en YouTube', 'Subir video', 'Publicar Shorts', etc.

Skill para crear notebooks marimo como archivos Python en el formato correcto. Se activa con solicitudes como 'crear un notebook marimo', 'notebook interactivo', 'notebook de Python', etc.

Habilidad para diseñar y optimizar secuencias de correo electrónico, campañas de goteo y correos escalonados. Se activa con solicitudes como 'diseña una secuencia de correo,' 'crea una campaña de goteo,' 'correo de bienvenida,' etc. Para onboarding dentro de la aplicación, consulte onboarding-cro.

Habilidad para crear PRDs (Documentos de Requisitos de Producto), elaborar especificaciones de funcionalidades y definir criterios de aceptación. Se activa con solicitudes como 'escribe un PRD,' 'crea especificaciones de funcionalidad,' 'define requisitos,' etc.

Habilidad para apoyar el diseno e implementacion de pruebas A/B y experimentos. Se activa con solicitudes como 'disenar una prueba A/B', 'quiero hacer un split test', 'configurar una prueba de hipotesis', 'comparar variantes', etc. Para la implementacion del seguimiento, consulte analytics-tracking.

Habilidad que genera automáticamente diagnósticos de errores o tutoriales de operación a partir de capturas de pantalla. Se activa con solicitudes como 'Analizar esta captura', 'Verificar el error de pantalla', 'Crear pasos de operación'.

Habilidad integrada para extraer elementos accionables y tareas del correo electronico y Slack. Utiliza Gemini 3.0 Flash para analisis de contexto, generando prioridades y borradores de respuesta. Se activa con solicitudes como 'revisar bandeja de entrada', 'verificar TODOs', 'mensajes por responder', 'revisar correo', etc.

1行の説明文

Habilidad para enrutar comandos slash de ai-agent-camp en Codex. Se activa con solicitudes como 'ejecutar /start-0-1', 'quiero usar comandos slash', 'enrutamiento de comandos', 'usar comandos de Cursor', etc.

Full lifecycle team skill — plan, develop, test, review in one coordinated session. Role-based architecture with coordinator-driven beat model. Triggers on "team lifecycle v4".

Iterative skill tuning via execute-evaluate-improve feedback loop. Uses ccw cli Claude to execute skill, Gemini to evaluate quality, and Agent to apply improvements. Iterates until quality threshold or max iterations. Triggers on "skill iter tune", "iterative skill tuning", "tune skill".

Universal team coordination skill with dynamic role generation. Uses team-worker agent architecture with role-spec files. Only coordinator is built-in -- all worker roles are generated at runtime as role-specs and spawned via team-worker agent. Beat/cadence model for orchestration. Triggers on "Team Coordinate ".

Unified team skill for tech debt identification and remediation. Scans codebase for tech debt, assesses severity, plans and executes fixes with validation. Uses team-worker agent architecture with roles/ for domain logic. Coordinator orchestrates pipeline, workers are team-worker agents. Triggers on "team tech debt".

Swarm intelligence team skill — ACO-driven multi-agent exploration with hybrid LLM coordinator + Python optimization controller. Coordinator generates swarm-config from user task, then runs K iterations of N parallel ants guided by pheromone state. Universal task space via config (nodes + scoring rule). Triggers on "team swarm", "swarm intelligence", "蚁群".

Unified team skill for code review. 3-role pipeline: scanner, reviewer, fixer. Triggers on team-review.

Unified team skill for testing team. Progressive test coverage through Generator-Critic loops, shared memory, and dynamic layer selection. Triggers on "team testing".

Unified team skill for quality assurance. Full closed-loop QA combining issue discovery and software testing. Triggers on "team quality-assurance", "team qa".

Guided reading of code or wiki to extract patterns

Use when exploring ideas, evaluating approaches, or needing multi-perspective analysis before implementation

Load and execute tool specs by category or name

Bump version, generate changelog, tag milestone

Use when a topic needs structured multi-dimensional investigation before planning or decision-making

Show project dashboard with progress and next steps

Get alternative perspectives — review, challenge, or consult

Discover issues via multi-perspective analysis

Detect version, preview changes, apply workflow upgrades

Capture and search learning insights and tips

Generate overlays to fix workflow command deficiencies

Refresh codebase docs from recent changes

Use when accumulated tech debt needs systematic identification and safe reduction

Initialize specs from project structure

Extract design system from code, generate reference package, persist as knowledge assets

Manage editing boundary restrictions