Loading...
Loading...
Found 1,837 Skills
Build structured hierarchical memory systems for LLM agents using GAM (General Agentic Memory) with support for text, video, and agent trajectories
Multi-agent deep research for comprehensive market analysis using the aipa CLI. Use this skill when the user asks for deep research, thorough market analysis, sector-wide investigation, comprehensive stock comparison, or detailed financial report. This runs a supervisor → parallel workers → aggregator → reviewer pipeline that takes longer but produces more thorough results than a simple analyze. Trigger for requests like "research banking sector", "deep dive into real estate stocks", or "comprehensive market overview". Can also incorporate fundamental analysis (PE, ROE, NPL, CAR, financial ratios) via `aipa fundamentals` when the user asks for fundamental context alongside technical research.
Lets end users add, authenticate, and manage MCP servers from the browser in assistant-ui apps with @assistant-ui/react-mcp. Use when building user-managed MCP server UIs: mounting McpManagerResource via useAui({ mcp }), declaring presets with defineConnector, dropping in McpConfigDialog, or composing McpManagerPrimitive (Root, Connectors, CustomServers, AddCustomTrigger), McpServerPrimitive (Root, Name, Icon, Status, ConnectButton, DisconnectButton, OAuthLink, RemoveButton, Error), and McpAddFormPrimitive (NameField, UrlField, AuthSelect, AuthFields, Submit, Cancel). Covers auth modes none/bearer/oauth, the OAuth flow with McpOAuthCallback, connection states, storage via McpLocalStorage/McpMemoryStorage/McpCustomStorage, reading state with useAuiState (s.mcp, s.mcpServer), and imperative addCustomServer/connect/callTool. Distinct from developer-defined backend @ai-sdk/mcp tools in the tools skill. Reach for this when connected-server tools are missing, OAuth never completes, or servers do not persist.
Audit a live page for accessibility issues and locate each violation precisely — optionally pass a URL (e.g. `accesslint:scan https://example.com/dashboard`), otherwise ask for one. Ensures a debuggable Chrome, runs the @accesslint/core engine via CDP, and returns a worklist of live-DOM WCAG violations grounded to each violation's DOM selector and source file:line. Locates; doesn't edit — output drives fixes by Claude. Use it for "is this page accessible", or to verify a UI change. For diffing against uncommitted changes or a branch, use the `diff` skill.
Grounding DINO for open-set object detection. Combines DINO-style detection with a BERT text encoder for language-guided detection — detects objects described by text prompts without a fixed class vocabulary. Use when training, evaluating, exporting, quantizing, or running inference for a TAO Grounding DINO model. Trigger phrases include "train Grounding DINO", "open-vocabulary detection", "text-prompted detector", "language-guided object detection".
OCDNet for scene text detection. Detects arbitrary-oriented text regions in natural images using a differentiable binarization approach. Use when training, evaluating, exporting, pruning, quantizing, retraining, or running inference for a TAO OCDNet model. Trigger phrases include "train OCDNet", "scene text detection", "arbitrary-oriented text boxes", "differentiable binarization detector".
Run OpenMMDL molecular dynamics workflows via the FastFold Workflows API (`openmmdl_v1`) from local topology + optional ligand files, prepare draft scripts, execute drafts, wait for completion, fetch artifacts/metrics, and extract trajectory frames. Use when users ask for OpenMMDL, protein-ligand MD, OpenMMDL script preparation, or `/openmmdl/results/<workflow_id>` reruns.
Run molecular dynamics (MD) simulations via the FastFold Workflows API. Today supports the CALVADOS+OpenMM workflow (calvados_openmm_v1) from either an existing fold job (AF structure + PAE auto-resolved) or manual PDB+PAE upload, then waits for completion, fetches metrics/plots/CSV artifacts, and extracts trajectory frames as PDB files. Use when running an MD simulation with FastFold, CALVADOS + OpenMM, reading MD metrics/plots, extracting frames, or scripting submit → wait → results for an MD run.
Static source-code vulnerability scan. Reads a target directory (and THREAT_MODEL.md if present), spawns parallel review subagents per focus area, and writes VULN-FINDINGS.json + .md for /triage to consume. Read-only — no building, running, or network. For execution-verified crashes, use vuln-pipeline instead. Use when asked to "scan for vulns", "review this code for security issues", "find bugs in <dir>", or as the step between /threat-model and /triage.
Scan the host repo a set of agent skills is installed into and reconcile every installed skill's config.json with detected facts — base branch, monorepo package roots, changelog directory, Linear issue-key prefixes, review bots, protected branches — plus the Linear team name and workspace slug fetched via the Linear MCP. Use when first installing these skills into a repo, or to refresh the configs after the skill set or repo layout changes. Also emits a committed `.claude/skills.lock` inventory of installed skill versions, and ensures the preflight skill's `.preflight-summary.json` scratch output is gitignored. Idempotent and safe to re-run: it reconciles drift rather than clobbering deliberate manual edits, presents a dry-run diff first, and only writes after confirmation — preserving each config's key order and formatting so a no-op run leaves files byte-identical.
Discovers requirements, and generates architectural, design, and deployment guidance for a retrieval-augmented generation (RAG)-capable enterprise search system in Google Cloud. Use when users need a vector-enabled SQL database as the store and index for the embedding vectors, an open model and open-source inferencing framework, and Kubernetes containers to host all the application components. DON'T use this skill for fully-managed RAG, or SaaS search services, or when a non-SQL vector database is required.
Define a precise React Doctor rule contract before implementation. Use when validating a rule idea, collecting official or open-source evidence, identifying false-positive traps, choosing detector precision, or setting first-version boundaries.