Find plugins

by Michael Kissinger

Spec-anchored workflow for autonomous coding runs on GPT-6 Astra in Codex: brief, run, verify, checkpoint. astra mode has Astra implement each slice; astra-crew routes implementation to gpt-5.6-sol or gpt-5.6-luna subagents by task class. Same spec contract as the Claude Code plugin in this repo.

by Michael Kissinger

Spec-anchored workflow for autonomous coding runs on Claude Fable 5.1: brief, run, verify, checkpoint, with a fable-crew provider mode that routes implementation to gpt-5.6-sol or gpt-5.6-luna. A sibling Codex plugin in codex/ runs the same spec on GPT-6 Astra. Built from Anthropic's Fable 5.1 docs.

by Clownware

A decision layer for Claude Code hooks. Classifies proposed tool calls against a policy you own and returns allow / ask / deny, using a fast System One model instead of a frontier LLM. Observe-only until you have looked at your own calibration table.

by Mason Beard

Shared Git-backed knowledge, session continuity, graph context, and wiki maintenance for coding agents.

by Ivan-WG

A committed, Obsidian-style codebase map for a project: brain/INDEX.md indexes areas, key files, and notable classes/entrypoints so a fresh Claude session starts oriented. A SessionStart hook injects the compact map (with a staleness hint when it lags HEAD); /brain reads it and /brain index builds/refreshes it. Builds the map only — it never annotates code, and nothing outside brain/ is ever modified.

by Aleck Yann

Second brain in OKF markdown, curated by Claude through pull requests: validator, pull request loop, Stop and SessionStart hooks, and skills to set up, capture, curate, ask, lint, review stale notes, approve, and fill the weekly rhythm table from the calendar.

by Joanna Seymenarova

Build a live status page for a repository's recent branches — what exists, who moved it, how far it got, what it touches, and what its own plan documents say the goal is.

by Engineering

Publication-safety review for personal-brand content: the legal-reviewer subagent checks a draft post, article or comment for employment-contract, NDA, defamation and disclosure risk, plus the employer-safety filter it marks against. Publishing skills are not part of this plugin yet.

by Jung Jiwon & Lee Jongkwan

The brand-design role on contract v3: 브랜드 정체성이 시각적으로 일관되는가. use_when: 브랜드 자산 신설/변경이 걸릴 때. Hand-off: 토큰 시스템화 구현은 → ux-engineering.

by clssck

Restate the assistant's previous response in plain, concise language.

by BrowserClaw Team

Control your own browser with Agent. High-efficiency Chrome browser automation MCP server and AI agent skill.

by Q

Drive Firefox, Zen, and Safari from the terminal — open pages, read and drive them, and take screenshots. Firefox and Zen run as isolated headless instances over Marionette; Safari runs in its own automation window over safaridriver.

by Buaflow

Buaflow workflow for Claude Code: intent → elaborate → spec → plan → task → check → done, with guard hooks. The gate and evidence controls stay committed in the project (.claude/), because pre-push and CI run them outside any session.

by Tyrone Ross

Multi-phase build orchestration with native debugging: assess, plan, execute, review, iterate. Plan-verify gate on Phase 2. Optional UI design gates. Native debug-loop, debugging-memory, and logging-tracer skills provide root-cause investigation without a bundled MCP server. Orchestrator owns when-to-fire; bundled skills own procedural detail; bridges are optional escalation hops to supporting plugins for extended capability.

by BulkTranscripts

YouTube transcripts for one video, a whole channel or a playlist, plus YouTube search and free new-upload tracking, from the hosted BulkTranscripts MCP server (OAuth sign-in, 30 free credits).

Bundles its own MCP server source. Includes 1 MCP server.

by Fixture Author

Renders a burndown of open findings as one static page. Writes only the page it is asked to write.

by epicsagas

BuildYourOwnHarness — profile → compile → install → evolve a personalized AI agent harness. Synthesis recombines vetted skills/agents by keyword under 3 safety gates.

by Guy Lifshitz

Phased, gate-enforced build pipeline for AI code generation. TDD mandatory, reviews scored, skipping blocked.

by SamMausberg

Write, check, test and tune CAIRN programs: the language skill, the cairn command on PATH, the language server, so every edit to a .cairn file comes back with the compiler's diagnostics, and an MCP server with the compiler's guarded edit, plan and implementation hosts.

by Candle

Candle launchpad skills: launch and seed tokens, trade with structural custody safety, read markets with no API key, react to events via webhooks, keep an automated strategy's books honest, and research strategies that can actually be rejected

by Ivan-WG

Candour as a mechanism, not a pep talk. One blocking Stop gate, five clauses a script can prove: a file:line citation resolving to no file or past its end; a position retracted after bare pushback with no tool call in between; a completion claim after edits with nothing executed since; a task-runner run stopping without its recorded gate pass; a dependency map changed with an untouched lockfile. Flattery, apologies and defensive phrasing are counted by /candor:check and blocked by nothing, on purpose. Plus the terse reply mode (/candor:level lite|full|ultra|off): chat-message brevity as a shape contract — prose-line budgets and a fixed report skeleton, never less work in the turn.

by Kyle Tabor

Corrective and Preventive Action (CAPA) process skill. Investigates process failures, researches best practices, fixes the process first, then re-executes to fix the original defect. Includes SQLite tracking database for pattern analysis across incidents.

by Jung Jiwon & Lee Jongkwan

The capacity-planning role on contract v3: 향후 수요 성장 대비 자원이 충분하며 언제 증설해야 하는가. use_when: 용량 예측/증설 시점 결정이 걸릴 때. Hand-off: 성능 자체의 병목 원인 분석은 → performance-engineering.