brainstorming-gatekeeper
Pre-flight check that determines if brainstorming is required before any action. Invoke this FIRST for any request involving creative or generative work.
Skills
Find an Agent Skill for your task. Read the guide and copy the install command.
Updated 11-06 00:14
Agents & models
25,712 results
dispatching-parallel-agents
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
exploration
Guide AI agents through oil & gas exploration workflow.
midstream
Guide AI agents through pipeline transportation and storage operations.
refining
Guide AI agents through refinery operations and process optimization.
using-petropowers
Use when starting any conversation - establishes how to find and use skills, requiring Skill tool invocation before ANY response including clarifying questions
kweaver-core
Operate KWeaver knowledge networks and Decision Agent — build knowledge networks, query Schema/instances, semantic search, execute Actions, Agent CRUD and conversations, Trace data analysis. Operate the Skill management module — register Skills, marketplace lookup, progressive reading, download and install. Operate the Vega observability platform — query…
bkn-ontology-builder
Use when a business-reviewable KWeaver BKN ontology modeling proposal needs to be generated, revised, or compared based on a PRD, meeting minutes, process descriptions, system/data materials, a BKN_Creator handoff summary, or an existing proposal.
kweaver
Operate the KWeaver knowledge network and Decision Agent — build knowledge networks, query schemas/instances, semantic search, execute Actions, Agent CRUD and conversations. Operate the Vega observability platform — query catalogs/resources/connector types, health inspections. When the user mentions "knowledge network", "knowledge graph", "query object classes", "execute...
agent-desktop-windows
Windows platform guide for the agent-desktop CLI. Lists per command group what works on Windows and what refuses, the first-contact gotchas (PowerShell ref…
langchain agent starter kit
ALWAYS START HERE for any LangChain, Deep Agents, or LangGraph open source agent project. It is the required starting point for any LangChain open source agent…
langfuse-onboarding
Configure an agent for whoever is using it — outside contributor or Langfuse maintainer, which areas they work on, and whether Linear answers. On Cursor Cloud,…
autoloop-controller
When continuous automated improvement of a Skill is needed. Wraps improvement-orchestrator in a persistent loop with convergence detection…
improvement-discriminator
Use when you need multi-reviewer blind scoring of improvement candidates, semantic evaluation with an LLM, to determine whether a candidate should be accepted, or when all scores are hold and you want to know why. Supports --panel multi-reviewer blind review and --llm-judge semantic evaluation. Not for structural evaluation (use improvement-learner) or gate decisions (use improvement-gate).
improvement-evaluator
Use when it is necessary to verify whether a Skill improvement truly improves AI execution effectiveness. Run AI tasks through a predefined task set (YAML), determine pass/fail, and output execution_pass_rate. Not used for document structure scoring (use improvement-learner) or candidate scoring (use improvement-discriminator).
improvement-executor
Use when approved improvement candidates need to be applied to target files, previous changes need to be rolled back, or change effects need to be previewed. Supports 4 actions (append/replace/insert_before/update_yaml), with automatic backup before each change. Not used for scoring (use improvement-discriminator) or gate validation (use...
improvement-generator
Use when you need to generate improvement candidates for a target skill, inject the previous failure information into the next round of generation, or analyze historical memory patterns to avoid repeated failures. Supports --trace to inject failure context. Not for scoring (use improvement-discriminator) or evaluation (use improvement-learner).
improvement-orchestrator
Use when you need to run the full pipeline of "generate → score → evaluate → execute → gate" in one click, automatically retry after failure, or batch-improve multiple skills. Not for evaluating skill quality alone (use improvement-learner) or manual scoring (use improvement-discriminator).
prompt-hardening
Harden agent prompts, system prompts, SOUL.md, AGENTS.md, and cron prompts to make LLMs reliably follow instructions. Trigger words: agent not listening, ignoring rules, bypassing constraints, prompt optimization, instruction compliance, rule reinforcement, prompt hardening, LLM not complying, model violations, creative...
session-feedback-analyzer
Parse Claude Code session JSONL to extract implicit user feedback signals. Detects skill invocations (tool_use blocks with name="Skill" or…
skill-distill
Use when multiple skills with overlapping functionality need to be merged into one distilled skill. Not applicable to extracting rules from skills (rules extraction is a separate capability from ECC's rules-distill skill, not part of this repo) or from...
skills-cli-sync
Sync skills-desktop's AGENT_DEFINITIONS and UNIVERSAL_AGENT_IDS with the upstream vercel-labs/skills CLI, then update docs and validate. Use when bumping…
legal-risk-analysis-goodlegal
Use for Legal Risk Analysis with Goodlegal MCP
recherche-jurisprudence
Search and cite French case law and statutes with the LibreJustice MCP tools. Use whenever the user asks about French or European court decisions (Cour de…