AgentEvolver

Create, optimize, and evaluate reusable AgentEvolver HTML Workflows. Use when a repeated successful multi-agent orchestration should be retained, when an evolvable Workflow has recurring orchestration-level defects, or when a live version needs evidence for keep, improve, or rollback decisions.

Create new skills, improve/optimize existing skills, and evaluate skill quality — the full skill lifecycle. Use this whenever the task involves authoring a skill from scratch, editing or improving an existing skill, evaluating/scoring a skill, running test cases to measure whether a skill helps, or tuning a skill's description for better triggering. MetaAgent uses it to orchestrate the create→evaluate→improve loop across sub-agents.

Create new agents, improve/optimize existing agents (both Python class and HTML prompt), and evaluate agent quality — the full agent lifecycle in this framework. Use whenever the task involves authoring a new agent, editing/improving an existing agent or its prompt, or evaluating/scoring an agent. MetaAgent uses it to orchestrate the create→evaluate→improve loop across sub-agents.

Create new connectors (CONNECTOR.md configs that connect the framework to an MCP server), improve/optimize existing connectors, evaluate connector quality, and — when needed — build a new MCP server to connect to. Use this whenever the task involves adding a connector for an MCP server, writing/editing a CONNECTOR.md, evaluating whether a connector's actions let an agent accomplish tasks, or building an MCP server from scratch (Python FastMCP or Node/TS). MetaAgent uses it to orchestrate the create→evaluate→improve loop across sub-agents.

Create new environments, improve/optimize existing environments, and evaluate environment quality — the full environment lifecycle in this framework. Use whenever the task involves authoring a new environment (a stateful Python class exposing actions, plus an ENVIRONMENT.md), editing/improving an existing environment, or evaluating/scoring one. MetaAgent uses it to orchestrate the create→evaluate→improve loop across sub-agents.

Create new tools, improve/optimize existing tools, and evaluate tool quality — the full tool lifecycle in this framework. Use whenever the task involves authoring a new tool (a Python class the agents can call), editing/improving an existing tool, or evaluating/scoring a tool. MetaAgent uses it to orchestrate the create→evaluate→improve loop across sub-agents.

Predict genome-wide functional tracks (RNA-seq, CAGE, DNase, ChIP) from DNA sequence with Borzoi. Use this skill when: (1) Scoring the regulatory effect of a variant on expression/accessibility, (2) Generating predicted coverage tracks for a locus, (3) Prioritising non-coding variants by predicted track delta.

Embed proteins with Meta AI's ESM-2 (`fair-esm` package). Use this skill when: (1) Extracting per-residue or per-sequence embeddings for downstream ML, (2) Masked-LM likelihood / mutation effect scoring, (3) Contact prediction from a sequence.

Score, embed, and generate DNA sequences with Evo 2, a long-context genomic foundation model. Use this skill when: (1) Computing per-nucleotide or per-sequence likelihoods for variant effect scoring, (2) Embedding genomic windows for downstream classification, (3) Generating DNA conditioned on a prefix, (4) Scoring regulatory or coding regions across species.

Generate a therapeutic indication dossier. Covers the patient population, epidemiology, disease biology, standard of care, regulatory precedent, and landmark clinical trials.

Embed and annotate single-cell expression data with scGPT, a foundation model for single-cell biology. Use this skill when: (1) Producing cell embeddings from an AnnData for clustering/integration, (2) Zero-shot or fine-tuned cell-type annotation, (3) Gene-level representation for perturbation/GRN tasks. For probabilistic single-cell models (scVI etc.), use the scvi-tools library.

Probabilistic single-cell RNA-seq with scvi-tools — scVI for a batch-corrected latent space, scANVI for semi-supervised label transfer, and Bayesian differential expression. Reach for this skill to integrate scRNA-seq batches, embed cells for clustering, transfer annotations from a reference onto a query, or score differentially expressed genes per cluster. For spatial deconvolution / mapping use the cell2location, DestVI, or Tangram methods instead.

Design guidance and fundamentals for building polished self-contained HTML/web artifacts (pages, dashboards, tools, docs). Use when generating an HTML page or web UI and you want a deliberate, non-templated visual identity calibrated to the task.

Structured workflow for authoring a substantial written document — proposal, technical spec, decision doc, PRD, RFC, design doc, or similar prose-heavy deliverable. Gather context, draft section by section, then reader-test the draft to catch blind spots before it ships. Use when the deliverable is a structured written document; for data/figure analysis reports use report_design_skill instead.

Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.

Use this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text/tables from PDFs, combining or merging multiple PDFs into one, splitting PDFs apart, rotating pages, adding watermarks, creating new PDFs, filling PDF forms, encrypting/decrypting PDFs, extracting images, and OCR on scanned PDFs to make them searchable. If the user mentions a .pdf file or asks to produce one, use this skill.

Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions "deck," "slides," "presentation," or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.

Produce a polished, self-contained analysis report (markdown or HTML) from findings and generated figures — data/numerical analysis reports, deep-research reports, deep-dive analysis reports, evaluation write-ups. Use whenever the deliverable is a REPORT that presents results together with charts/tables/images. Its core discipline — every figure is EMBEDDED inline (never just listed by filename), each figure sits under the claim it supports, and the finished report is verified to render.

Toolkit for styling artifacts with a theme. These artifacts can be slides, docs, reportings, HTML landing pages, etc. There are 10 pre-set themes with colors/fonts that you can apply to any artifact that has been creating, or can generate a new theme on-the-fly.

Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.

The global playbook for self-evolution — how MetaAgent improves the system's OWN capabilities (agents, prompts, tools, skills, connectors, environments) while serving a user task. Use whenever a task is blocked or degraded by a missing/weak capability, when a sub-agent repeatedly fails for a fixable reason, or when the user explicitly asks to create/improve/evaluate a capability. Provides the cross-cutting loop (decide → generate/optimize → evaluate → adopt or roll back), the enable_evolving gate rules, and how to drive the per-type creator skills. NOT for the user's own deliverable work.

Refines raw ideas into sharp, actionable concepts through structured divergent and convergent thinking. Use when an idea is still vague, when you need to stress-test assumptions before committing to a plan, or when you want to expand options before converging on one. Triggers on "ideate", "refine this idea", or "stress-test my plan".

Extracts what the user actually wants instead of what they think they should want. Achieves this through one-question-at-a-time interview until ~95% confidence about the underlying intent. Use when an ask is underspecified ("build me X" without "for whom" or "why now"), when the user explicitly invokes ("interview me", "grill me", "are we sure?", "stress-test my thinking"), or when you catch yourself silently filling in ambiguous requirements before any plan, spec, or code exists.

Guides stable API and interface design. Use when designing APIs, module boundaries, or any public interface. Use when creating REST or GraphQL endpoints, defining type contracts between modules, or establishing boundaries between frontend and backend.

Guides systematic root-cause debugging. Use when tests fail, builds break, behavior doesn't match expectations, or you encounter any unexpected error. Use when you need a systematic approach to finding and fixing the root cause rather than guessing.

Records decisions and documentation. Use when making architectural decisions, changing public APIs, shipping features, or when you need to record context that future engineers and agents will need to understand the codebase.

Structures git workflow practices. Use when making any code change. Use when committing, branching, resolving conflicts, or when you need to organize work across multiple parallel streams. Use when cutting a release, choosing a semantic version bump, tagging, or writing a changelog.

Delivers changes incrementally. Use when implementing any feature or change that touches more than one file. Use when you're about to write a large amount of code at once, or when a task feels too big to land in one step.

Grounds every implementation decision in official documentation. Use when you want authoritative, source-cited code free from outdated patterns. Use when building with any framework or library where correctness matters.

Drives development with tests. Use when implementing any logic, fixing any bug, or changing any behavior. Use when you need to prove that code works, when a bug report arrives, or when you're about to modify existing functionality.

A demonstration skill that greets users and showcases all skill elements. Use when the user asks for a greeting, wants to test the skill system, or requests a hello world example.

Initialize a new CLAUDE.md file documenting the codebase. Use when asked to create or improve a CLAUDE.md / agent-guidance file that future agent instances will read to operate in this repository.

Optimizes agent context setup. Use when starting a new session, when agent output quality degrades, when switching between tasks, or when you need to configure rules files and context for a project.

Subjects every non-trivial decision to a fresh-context adversarial review before it stands. Use when correctness matters more than speed, when working in unfamiliar code, when stakes are high (production, security-sensitive logic, irreversible operations), or any time a confident output would be cheaper to verify now than to debug later.

Breaks work into ordered tasks. Use when you have a spec or clear requirements and need to break work into implementable tasks. Use when a task feels too large to start, when you need to estimate scope, or when parallel work is possible.

Creates specs before coding. Use when starting a new project, feature, or significant change and no specification exists yet. Use when requirements are unclear, ambiguous, or only exist as a vague idea.

Deep research harness — fan-out web searches, fetch sources, adversarially verify claims, then synthesize a cited report. Use when the user wants a deep, multi-source, fact-checked research report. If the question is underspecified, ask 2-3 clarifying questions to narrow scope first.

Instruments code so production behavior is visible and diagnosable. Use when adding logging, metrics, tracing, or alerting. Use when shipping any feature that runs in production and you need evidence it works. Use when production issues are reported but you can't tell what happened from the available data.

Review the current diff for correctness bugs and reuse/simplification/efficiency cleanups at a chosen effort level (low/medium → fewer, high-confidence findings; high → broader coverage, may include uncertain findings). Use when asked to review a diff, find bugs in a change, or check a branch before merge.

Optimizes application performance. Use when performance requirements exist, when you suspect performance regressions, or when Core Web Vitals or load times need improvement. Use when profiling reveals bottlenecks that need fixing.