meta-cognition
Meta-cognitive framing for analyze-before-doing, ownership routing, risk gating, minimum-closure planning, and retrospective extraction in multi-agent work.…
Skills
Find an Agent Skill for your task. Read the guide and copy the install command.
Updated 11-05 00:16
Agents & models
25,666 results
ml-engineer
Build production ML systems with PyTorch 2.x, TensorFlow, and modern ML frameworks. Implements model serving, feature engineering, A/B testing, and monitoring.…
moltbook
Interact with Moltbook social network for AI agents. Post, reply, browse, and analyze engagement. Use when the user wants to engage with Moltbook, check their…
self-improving agent (with self-reflection)
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use before…
skills-search
Search skills.sh registry from CLI. Find and discover agent skills from the skills.sh ecosystem.
task-supervisor
Periodically inspect the execution status of agent tasks, identify anomalies such as being stuck, aborted, or producing no output, and trigger reminders or alerts.
tcm-meridian-inference
TCM meridian inference engine — health scoring from 6-meridian measurements
thinking-jensen-huang
A practical framework that distills Jensen Huang's thinking model. Activate it when you need to accelerate computing vision, ecosystem platform thinking, or long-term technology bet-style thinking.
prd-to-kandev-tasks
Break PRDs, specs, or feature briefs into Kandev task graphs and submit them through MCP or agentctl. Use when asked to decompose product work, create Kandev…
prd-todo-slicer
Break a PRD.md into a small, deterministic TODO.md queue and per-task briefs that sub-agents can execute safely one task at a time.
abel
Use when the user asks for Abel or starts an Abel workflow and you need to check auth state, initialize Abel if needed, and route to the right Abel skill…
abel-ask
Use when the user wants an Abel causal map, graph exploration, target or candidate discovery, company/market mechanism read, or life/business investment…
bootstrap-cap-server
Use when Codex needs to turn an existing causal graph, causal runtime, or internal causal API into a CAP-compatible server, especially when the honest CAP…
goalify
Hand Claude a huge task. Come back to proof it's done — not a promise that it is. goalify writes the brief (a file with the instructions the run works from)…
capture-feedback
Capture concise user corrections about how an agent should have behaved differently, saving a local shared feedback note under ~/.agents/capture-feedback for…
broker-vocabulary
Vendor-neutral vocabulary for programmatic broker integration: the five access archetypes, the second axis separating a single broker from a multi-broker…
cross-model-peer-review
Doctrine for putting a plan, a spec, or a session's own decisions in front of a second model family: when the cost is earned, the GIVEN versus DERIVED…
prompt-engineering
Knowledge base behind the prompt-engineer agent and /prompt-optimize: the source-of-truth order for model facts, the model-class gate, and six on-demand…
deeplake-plugin
Cloud-backed shared memory for AI agents powered by DeepLake. Install once, memory persists across sessions, machines, and channels.
mcp-builder
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when…
call-adcp-agent
Wire-level invariants for any AdCP buyer call — idempotency_key replay semantics, account `oneOf` variants, async `status:'submitted'`+`task_id` polling, error…
melech
Remote-first catalog of AdirD/agent-shell-hamelech ("melech") skills: for each remote skill show name, description, installed y/n, where (global / project /…
agentos
Trigger this skill when the user expresses intent to generate, create, draw, or render an AI image, illustration, picture, or artwork. Payment is handled…
get-started
First-run setup for the Aeon plugin. Checks the connection to the user's Aeon agent, shows what it can do, and runs one skill so the user sees it work. Use…