LLM Mart Basic

Docs specialist — docstrings, API refs, README, standalone FAQ/comparison tables. NOT for CHANGELOG (oss:shepherd), linting (foundry:linting-expert), implementation (foundry:sw-engineer), narrative content (foundry:creator). TRIGGER: "write docs for", "add docstrings to", "update

Python static analysis — ruff, mypy, pre-commit, lint/type fixes, type annotations. NOT for CI topology (oss:cicd-steward), test logic (foundry:qa-specialist), non-style implementation (foundry:sw-engineer), docstrings (foundry:doc-scribe). TRIGGER: "is this clean", "lint issues"

Perf engineer — CPU/GPU/memory/I/O bottlenecks, DataLoader throughput, PyTorch tuning. Profile-first, measures before changing. NOT for refactoring (foundry:sw-engineer), architecture (foundry:solution-architect), DataLoader correctness (research:data-steward). TRIGGER: "why is t

QA specialist writing/fixing tests. Black-box tester: public API surface, expectations from docs not implementation. NOT for linting (foundry:linting-expert), implementation (foundry:sw-engineer), test perf (foundry:perf-optimizer), non-Python frameworks. TRIGGER: "write tests fo

Architectural spec specialist — ADRs, API design, migration plans, component diagrams. Reads code, produces specs only. NOT for implementation (foundry:sw-engineer), release mgmt (oss:shepherd), adversarial challenge (foundry:challenger), perf tuning (foundry:perf-optimizer). TRI

Senior SW engineer writing/refactoring Python — features, bugfixes, TDD, SOLID. Also authors hook JS files under hooks/. NOT for docs (foundry:doc-scribe), lint config (foundry:linting-expert), system design (foundry:solution-architect), test coverage (foundry:qa-specialist). TRI

Fetches web pages, API docs, external package/release info — version lookups, GitHub release extraction, docs scraping. NOT for code analysis (foundry:sw-engineer), ML paper analysis (research:scientist), internal docs (foundry:doc-scribe), local codebase search. TRIGGER: "check

CI/CD health specialist, Python/GitHub Actions only — failing CI runs, build times, test matrices, caching, SHA pinning. NOT for ruff/mypy config (foundry:linting-expert), PyPI release/CHANGELOG (oss:shepherd), non-GitHub-Actions platforms. TRIGGER: failing CI runs, slow builds,

Fetches all GitHub API data for a repo (REST + GraphQL) in two parallel groups; writes raw JSONL for oss:repo-warden axis scorers. TRIGGER when: spawned by /oss:analyse (vitality mode) to fetch raw GitHub data. NOT for axis scoring or report generation. NOT for direct user invoca

Scores an assigned group of vitality axes from a pre-fetched DATA_FILE using vitality-scoring.md; writes partial scores JSON for /oss:analyse assembly. TRIGGER when: spawned 3× in parallel by /oss:analyse (vitality mode) to score axis groups A, B, or C. NOT for raw data fetching

OSS shepherd, Python/ML/CV/AI — contributor communication (triage, reply/PR drafts), release coordination (SemVer, PyPI, CHANGELOG). NOT for docstrings/README (foundry:doc-scribe), CI/publish YAML (oss:cicd-steward), diff review (/oss:review), CHANGELOG gen (/oss:release). TRIGGE

Data lifecycle specialist — dataset acquisition, DVC versioning, split audits, leakage detection, DataLoader config. Manual invocation only — no research skill auto-dispatches this agent. Delegates scraping to foundry:web-explorer. NOT for ML experiment design (research:scientist

AI/ML researcher — paper analysis, hypothesis generation, experiment design. ONLY for named research paper/hypothesis/experiment. NOT for general Python (foundry:sw-engineer), SOTA surveys (/research:topic), web content (foundry:web-explorer), dataset acquisition (research:data-s

Creates original editorial illustrations where a recurring mascot character performs the idea — one caught scene by default, a hand-built explainer diagram (labeled stages, a fan-out, timeline, loop, or stack) when the structure itself is the point, or a transparent character cut

Use when the user has agreed to install the Hedgehog build discipline in a project that does not have it yet, or when they mention Hedgehog by name — carries the `npx @skyf0xx/hedgehog init` install procedure and its core and host flags. The plugin's SessionStart hook decides whe

Maintainer-only. Use when re-vendoring vendor-skills/BMAD/ against a newer BMAD-METHOD commit — "update BMAD", "re-vendor BMAD", "bump the BMAD pin". Not part of the Hedgehog discipline a consuming project copies; this only applies to the Hedgehog repo itself.

Maintainer-only. Use when triaging inbound GitHub issues and pull requests on skyf0xx/hedgehog — "triage the issues", "check the PRs", "review inbound", "what's in the queue". Reads each item read-only, judges it for security and for whether it is real, then fixes and closes or c

Use when uncommitted changes need to be split into atomic, conventional commits ordered for review. Triggers on "commit this", "make commits", "clean up commits", "commit the changes". In Hedgehog, each Loop step is already meant to be its own commit — this skill matters most whe

Use when the user wants to contribute a fix or ROADMAP.md item back to the Hedgehog project itself (skyf0xx/hedgehog) rather than their own project. Triggers on "let's fix that in Hedgehog", "I want to contribute", "let's pick up a roadmap item", or when `tweaker` offers this at

Use on any core for first-run planning intake — Phase 0 runs the vendored BMAD-METHOD planning shelf, shared by every core, and Phase 1 (mining `04-prd.md` into intent records plus the Add-ons/sync-and-remote-entities decision) is full-stack-app's and pwa-app's shared procedure —

Fourteen posts of being wrong in production, compressed to checkboxes

Healthy nodes, a quiet network, 300 restarts in three days, and a latency budget measured in milliseconds

Discovery worked. Ping worked. Every TCP connection timed out, and later the tunnel only worked when someone had a terminal open.

Every VM came back. The cluster did not. Declarative systems converge on config, and the datapath isn't config.

A surprising share of AI-in-the-terminal failures aren't the AI. They're zsh, and a version of bash from 2006.

A Claude Code plugin turns standalone project configuration into a namespaced, installable extension that teams and communities can update as one unit.

None of the safety came from the model. It came from six boring habits.

Skills package instructions and references. Subagents run work in a separate context and return results. They solve different problems and can be composed deliberately.

Six hours in, one step left, everything green, and the incident that didn't happen

CLAUDE.md carries persistent project context. Skills load reusable procedures when relevant. Separating stable facts from task-specific workflows keeps both easier to maintain.

Twenty minutes recovering secrets that never existed, and the one sentence from a human that ended it

An API request routing a model's tool call through an approval gate to a remote MCP server

31 config keys, two audits, and why the first one was wrong in both directions

The official MCP Registry stores standardized server metadata rather than package code. Publishers verify a namespace, describe installation or remote access, and submit immutable versions.

Everyone looks at the Dockerfile. The file that actually leaked the key was the project file.

Remote MCP authorization uses established OAuth standards, but secure integration still requires issuer validation, least-privilege scopes, protected token handling, and server-side enforcement.

"Copy it over and switch the reference" is two steps, and the outage lives in the one nobody checks

stdio fits local processes and prototypes. Streamable HTTP fits hosted services and shared integrations. The right choice follows where the capability runs and who must reach it.

The most important rule wasn't about what I could change. It was about what I was allowed to display.

Tools perform operations, resources expose readable context, and prompts provide reusable templates. Choosing the correct primitive makes an MCP server easier to understand and govern.

Run autonomous hunt loop on a target — scope check → recon → rank surface → hunt → validate → report with configurable checkpoints. Usage: /autopilot target.com [--paranoid|--normal|--yolo]

Build an exploit chain — given bug A, finds B and C to combine for higher severity and payout. Knows common chain patterns: IDOR→ATO, SSRF→cloud metadata, XSS→ATO, open redirect→OAuth theft, S3→bundle→secret→OAuth. Usage: /chain

Active vulnerability hunting. Two-track dispatcher — asks Red Team vs WAPT, hands off to hunt-dispatch skill and sibling commands. Usage: /hunt target.com | /hunt *.target.com | /hunt targets.txt [--vuln-class X] [--source-code P] [--chrome]

On-demand intelligence fetch for a target — CVEs, disclosed reports, new features. Pulls NVD/GitHub-Advisory CVEs + bundled disclosed reports + hunt memory context. Usage: /intel target.com

Inspect or rotate the autopilot ledger JSONL files (findings.jsonl, negatives.jsonl). Caps file size and keeps N rotated backups so memory does not grow unbounded.

Pick up a previous hunt on a target — shows hunt history and untested surface from the autopilot ledger. Usage: /pickup target.com

Run full recon pipeline on a target — subdomain enum (Chaos API + subfinder), live host discovery (dnsx + httpx), URL crawl (katana + waybackurls + gau), gf pattern classification, nuclei scan. Outputs to recon/<target>/ directory. Usage: /recon target.com

Optional manual note on a target or the last confirmed finding. Capture is automatic during autopilot; this is for extra context. Usage: /remember

Write a submission-ready bug bounty report. Generates H1/Bugcrowd/Intigriti/Immunefi format with CVSS 3.1 score, proof of concept, impact statement, and remediation. Run /validate first. Usage: /report

Mandatory pre-flight scope check — verify an asset is in scope BEFORE any HTTP touch. Deterministic (deny-wins, default-deny) via engine/scope.py against the engagement's scope.md. Blocks out-of-scope testing. Usage: /scope <asset> [<asset> ...]

Show ranked attack surface for a target from its recon manifest + hunt memory. Deterministic backing is `cbh surface <target>` (reads recon/<target>/manifest.json); LLM layer adds ledger signal. Usage: /surface target.com

Meme coin and token security scan — checks for rug pull vectors (hidden mint, honeypot, fee manipulation, LP lock bypass, authority retention, bonding curve exploits, fake renounce, sandwich amplification). Manual 8-class grep audit (with an optional automated scanner if present). Usage: /token-scan <contract_path_or_dir> [--chain solana]

Quick 7-Question Gate triage on a finding before writing a report. Kills N/A submissions before they happen. Faster than /validate — for quick go/no-go decisions. Usage: /triage

Validate a finding — runs 7-Question Gate + 4-gate checklist. Kills weak findings before report writing. Prevents N/A submissions that hurt validity ratio. Usage: /validate

Smart contract security audit — runs through 10 bug class checklist (accounting desync, access control, incomplete path, off-by-one, oracle errors, ERC4626, reentrancy, flash loan, signature replay, proxy/upgrade). Applies pre-dive kill signals first. Generates Foundry PoC template for confirmed findings. Usage: /web3-audit <contract.sol>

Crabbox is a single CLI (`crabbox`). Commands are top-level, not nested under a

`crabbox actions` prepares a leased box from your repository's own GitHub

See [Runtime adapter stack](../features/runtime-adapter-stack.md) for the

`crabbox admin` groups trusted operator controls for coordinator-backed leases and the cloud resources behind them. Use it to inspect every lease the broker tracks, reconcile expired leases against live cloud state, force-release or delete a backing server, print provider IAM pol

`crabbox artifacts` turns a desktop lease into durable QA evidence: it collects

Unofficial KakaoTalk CLI and native MCP server for macOS — read, watch, and send messages via Accessibility automation.

Local-first AI action assistant for operators: memory, skills, tools, and permission gates to turn work into controlled action.

Agent-driven media library for your cloud drives (Quark 夸克 / 115 / 光鸭 GuangYa / 123网盘 / 天翼 Tianyi)

An open-source desktop AI agent built around Cyrene’s persona and powered by the self-developed Cyrene_Harness framework. It combines immersive character chat w…

在 Android 上运行的 AI 编程 Agent，内置 Linux 终端与代码编辑器，支持 MCP 协议扩展。

TencentDB Agent Memory is a team-level memory hub for AI Agents — turning conversations, docs, and code into four reusable memory assets (Chat Memory, Skill, LL…

Floe — a native iOS/iPadOS AI agent workspace for iPhone and iPad, built for private bring-your-own-key workflows.

Spring Boot AI Agent — an out-of-the-box solution that makes your app converse, remember, think, and act.

Give your AI agent eyes for PDFs — structured text, tables, OCR, visual evidence, and page-level citations via MCP. Native Rust, local-first.

A tiny Claude Code skill that keeps your prompt cache warm during idle sessions, so your next message reads from cache instead of paying full price.

Monet — Multi-engine mission control for coding agents (Claude Code and Codex today). Browse, search, and drive your agent sessions from a native desktop app.

A DottedSign MCP server that enables AI assistants (Claude, ChatGPT) to manage signing tasks, templates, and document status via natural language.

Курируемый handbook по Claude Code на русском: hooks, skills, CLAUDE.md шаблоны, MCP-серверы, кейсы.

👾 Open Computer Use – Open-Source Alternative to Codex Computer Use

An open-source AI companion that actually remembers you — runs entirely on your own Mac or server. Desktop pet · long-term memory · proactive companionship · mu…

Graphical management tool for Kubernetes on desktop and mobile.

Open-source desktop app for content creation, with an agent runtime and standalone CLI.

Connect Claude and ChatGPT to KDAN PDF — upload, edit, compress, protect, redact, and compare PDFs in chat.

Ekko Studio is a local-first AI workspace for multi-agent chat, coding, and visual workflows, available on desktop and the web.

Open-source native iPhone/iPad client for OpenAI Codex CLI and Claude Code — review diffs, approve actions, steer sessions, and manage Git remotely.