Collection · published
av/skills
A library of 26 skills for agent workflows, writing, and local AI. Choose a skill and follow the source README to install it by name.
In this collection
- agent-integration-testing
Apply when the user requests integration testing, feature validation, or test plan execution.
Source/agent-integration-testingin Cursor’s slash menu - anneal
Finds and fixes duplicated logic, unnecessary complexity, silent errors, and inconsistent conventions in AI-written code.
Source/annealin Cursor’s slash menu - article-factory
Create a researched long-form article from a topic prompt via an orchestrated pipeline - research agent (first-person sources, working-definition gate).
Source/article-factoryin Cursor’s slash menu - autoresearch
Runs repeated experiments and keeps changes that improve one measurable outcome.
Source/autoresearchin Cursor’s slash menu - boost-modules
Develop custom modules for [Harbor Boost](https://github.com/av/harbor/tree/main/boost), an optimizing LLM proxy.
Source/boost-modulesin Cursor’s slash menu - bugbash
Systematically explore and test any software project (CLI, API, Backend, Library, etc.) to find bugs, usability issues, and edge cases.
Source/bugbashin Cursor’s slash menu - bughunt
Finds bugs in a defined area, verifies and fixes them, then checks the fixes against project constraints.
Source/bughuntin Cursor’s slash menu - catalog-deslop
Catalog-then-fix slop removal on an existing prose draft.
Source/catalog-deslopin Cursor’s slash menu - discipline
Gives agents a set of rules for scoping work, tracking constraints, verifying claims, and recovering from failures.
Source/disciplinein Cursor’s slash menu - factual-communication
Rephrase content as a flat list of atomic facts in the style of av/facts fact sheets.
Source/factual-communicationin Cursor’s slash menu - ideate
Runs a timed brainstorming session in which independent agents propose, critique, and compare ideas.
Source/ideatein Cursor’s slash menu - improve-skill
Reviews an existing skill, makes one focused change, and has an independent reviewer check the result.
Source/improve-skillin Cursor’s slash menu - make-video
Create a HyperFrames video composition from a brief/script autonomously.
Source/make-videoin Cursor’s slash menu - mosaic-writing
Assemble a finished piece from human-written fragments via an orchestrated two-loop pipeline - an assembly loop (assembler + cold reviewer).
Source/mosaic-writingin Cursor’s slash menu - preact-buildless-frontend
Build-less ESM frontends that run directly in the browser without bundlers.
Source/preact-buildless-frontendin Cursor’s slash menu - pull-llamacpp-model
Downloads llama.cpp models and describes a CPU fallback for ROCm environments.
Source/pull-llamacpp-modelin Cursor’s slash menu - run-llms
Set up and run local LLMs with Harbor.
Source/run-llmsin Cursor’s slash menu - sandcastle
Orchestrate AI coding agents (Claude Code, Codex, OpenCode) in isolated sandboxes using the @ai-hero/sandcastle SDK.
Source/sandcastlein Cursor’s slash menu - superclaude
Set up and operate the Claude Code harness for large codebases.
Source/superclaudein Cursor’s slash menu - timeboxed-iterating
Carry out a task iteratively over a user-specified duration by dispatching subagents.
Source/timeboxed-iteratingin Cursor’s slash menu - timeframe-research
Builds a sourced research dossier by dividing a date range into year and month research passes.
Source/timeframe-researchin Cursor’s slash menu - tinygrad
Deep learning framework development with tinygrad - a minimal tensor library with autograd, JIT compilation, and multi-device support.
Source/tinygradin Cursor’s slash menu - turso-db
Shows how to install, configure, and use Turso DB, a SQLite-compatible database engine written in Rust.
Source/turso-dbin Cursor’s slash menu - use-coding-agents
Shows how to call installed coding agents as headless workers from a coordinating session.
Source/use-coding-agentsin Cursor’s slash menu - workgraph
Orchestrate work as a cyclic directed graph of subagent-executed nodes with transition criteria on edges, budgeted cycles, and per-node gates.
Source/workgraphin Cursor’s slash menu - workmachine
Carry out work as an explicit event-driven state machine executed by subagents.
Source/workmachinein Cursor’s slash menu