Find plugins

by chock-core

Block hallucinated or unknown dependencies before they enter the repo. Watches requirements.txt, pyproject.toml, package.json, and go.mod, and blocks any newly added dependency not present in the allowlist file. Opt-in: disabled by default because it requires a curated allowlist. Enable with `chock enable verify-dependency-exists` after populating .chock/dependency-allowlist.txt. Runs at commit and at agent tool-use (a manifest edit is judged against the file on disk, and at the turn's end against HEAD).

by chock-core

Gate MCP server configuration as protected content. A shell write to .mcp.json is refused unless every mcpServers entry on the line, parsed as JSON, matches a name+source pair on the allowlist -- an unlisted name blocks, and an allowed name whose command/args/url changed blocks too. `claude mcp add|add-json` is checked the same way (add-from-claude-desktop cannot be verified and is refused). The allowlist ships inside this guard's own script, protected like every policy's implementations/ source -- edit only with 'chock: approved-config-change'. The shell guard reads Claude Code's .mcp.json, best-effort (PreToolUse fails open on a crash; a write with no visible content fails closed). A script gate at commit and tool use parses every written MCP config (.mcp.json, .cursor/mcp.json, .vscode/mcp.json, claude_desktop_config.json, .gemini/settings.json, .codex/config.toml) against the same allowlist; only servers the change adds or alters are refused, at commit and tool use. No pragma. [Session-enforced by the PreToolUse hook under com.github.copilot/ in clients that read that namespace (documented for VS Code agent mode); a client that ignores it, as the Agent Plugins spec tells generic clients to, gets the advisory skill only. The hook needs git, a usable bash and a Python 3.11+ (python3, python or py, whichever actually runs). With no working Python it exits 2; without git or bash, fail-open clients allow silently and fail-closed clients refuse matched commands. If the guard itself crashes or times out, the hook asks for confirmation rather than allowing silently -- VS Code agent mode honours that ask and it overrides the client's own auto-approve.]

by chock-core

Gate MCP server configuration as protected content. A shell write to .mcp.json is refused unless every mcpServers entry on the line, parsed as JSON, matches a name+source pair on the allowlist -- an unlisted name blocks, and an allowed name whose command/args/url changed blocks too. `claude mcp add|add-json` is checked the same way (add-from-claude-desktop cannot be verified and is refused). The allowlist ships inside this guard's own script, protected like every policy's implementations/ source -- edit only with 'chock: approved-config-change'. The shell guard reads Claude Code's .mcp.json, best-effort (PreToolUse fails open on a crash; a write with no visible content fails closed). A script gate at commit and tool use parses every written MCP config (.mcp.json, .cursor/mcp.json, .vscode/mcp.json, claude_desktop_config.json, .gemini/settings.json, .codex/config.toml) against the same allowlist; only servers the change adds or alters are refused, at commit and tool use. No pragma. [Session-enforced via a PreToolUse hook; needs git, a usable bash and a Python 3.11+ (python3, python or py, whichever actually runs; the Windows Store stub is skipped). With no working Python the hook refuses (exit 2); without git or bash, fail-open clients allow silently and fail-closed clients refuse matched commands. If the guard itself crashes or times out, the hook asks for confirmation rather than allowing silently.]

by chock-core

Gate MCP server configuration as protected content. A shell write to .mcp.json is refused unless every mcpServers entry on the line, parsed as JSON, matches a name+source pair on the allowlist -- an unlisted name blocks, and an allowed name whose command/args/url changed blocks too. `claude mcp add|add-json` is checked the same way (add-from-claude-desktop cannot be verified and is refused). The allowlist ships inside this guard's own script, protected like every policy's implementations/ source -- edit only with 'chock: approved-config-change'. The shell guard reads Claude Code's .mcp.json, best-effort (PreToolUse fails open on a crash; a write with no visible content fails closed). A script gate at commit and tool use parses every written MCP config (.mcp.json, .cursor/mcp.json, .vscode/mcp.json, claude_desktop_config.json, .gemini/settings.json, .codex/config.toml) against the same allowlist; only servers the change adds or alters are refused, at commit and tool use. No pragma.

version-drift-manifests is an open-source Claude Code and Agent Plugins plugin indexed from GitHub. Review its manifest and source before installing.

version-drift-marketplace is an open-source Agent Plugins plugin indexed from GitHub. Review its manifest and source before installing.

by devantler-tech

Build a product by conversation alone — a plain-language build companion agent plus skills for needs-to-stack translation, an allowed-stack guardrail, and a jargon-free voice; for people with no technical background

by Anton

Helps the teacher run a vibe-coding course: an interview-driven roadmap builder and a beginner-friendly per-lesson writer that save Russian Markdown files into the Obsidian vault's Курс/ folder

by grzmol

Salesforce delivery harness for Claude Code: parallel multi-agent workflow, documentation-grounded Salesforce and fflib skills, deterministic hooks, and local plus post-deploy verification gates.

Bring an existing video over the language barrier: download it, transcribe, subtitle, and optionally re-voice it in Chinese.

Produce an original video from source material: planning, provider-agnostic voiceover, and 4K render polish.

by atman-33

Produce visual deliverables as self-contained HTML and hand them over as PDF: fixed 16:9 slide decks, inline-SVG diagrams on an editorial grammar, readable HTML documents and dashboards, and publication-quality PDF export for all of them — rendered entirely inside Claude Code. Install at user scope; optional.

by Bruno Volpato

Create and render theme-aware diagrams, charts, docs figures, and animated GIFs from TSX.

by xingchen20lj

VulnGate — a Codex-native source-audit pipeline for ANY codebase: libraries (parsing/serialization), web frameworks (Spring/Struts), middleware/servers (Tomcat/Jetty), logging libraries (Log4j/Logback), expression engines, message/RPC stacks, and applications. Runs the S1→S8 audit loop (attack-surface map → candidates → source audit → PoC matrix → novelty gate → CVSS → disclosure-ready report) with hard evidence gates (G0–G5) driven by the host agent. Aliases: @vulngate, @vg, 漏洞审计, 源码审计, 审计管线, vuln audit. Coverage-driven auditing: full source inventory, heuristic call graphs and flows, security control gaps, sibling differentials, candidate scheduling, and macOS/Electron/native application reconstruction.

by Mingyang Bao

Unattended server & long-running task supervision — health checks, anomaly detection, auto-repair, multi-channel alerting. One YAML config per project, zero code changes to adopt.

Expert on Wealthfolio — the open-source, private, local-first portfolio & net-worth tracker (desktop, iOS, self-hosted Docker). Concepts, CSV import, self-hosting, and addon development.

by AlphaBravo

Generate really good dynamic Workflow scripts on demand. /weave:make <task> classifies the task into an orchestration archetype (understand / design / review / research / migrate / generic), picks the right fan-out topology (pipeline by default, barrier only when a stage genuinely needs all prior results), and authors a self-contained Workflow script that obeys the Workflow tool's hard rules — pure-literal meta block, schema'd agent returns instead of text parsing, .filter(Boolean) after fan-out, adversarial-verify for findings, loop-until-dry / loop-until-budget for unknown-size work, isolation:'worktree' only for parallel file mutation, no Date.now/Math.random. Every script is validated with `node --check` and a 12-point self-audit before it is saved to .claude/workflows/ or run. Weave writes the orchestration so you don't freehand it.

by Paul Hinz / The Linux Foundation

LFX Marketing OS Website Designer Agent (Creative Services, Agent No. 9, Foundation Setup). Turns an LF project's three foundation docs into an approved sitemap, a complete Hugo website, and a GitHub-deployed, DNS-ready launch package — with four human approval gates.

by AlphaBravo

Production documentation for an app you built, in a fixed layout modelled on Harvester: subject-first directories and a Docusaurus site whose navigation is the filesystem. Every page declares both what it is and who it is for, with sections from The Good Docs Project and quality checks from the ISO/IEC/IEEE 26514 characteristics: type mixing, alt text, heading order, and reading grade measured against that page's own reader. Extracts the reference material, anchors every claim to a file and line, measures drift against a recorded manifest, and refuses to state what it could not check.

by gxgeek-n

浙江大学工程管理硕士(MEM)学位论文一体化写作技能包(21 skill)：从选题、文献、写作、语体、降AIGC、数据可视化、审稿、评审应对、LaTeX编译、Word导出、中期报告到答辩准备的全流程。以 zjuthesis 为参照，方法论可迁移到其它中文学位论文。

with-agents is an open-source Agent Plugins plugin indexed from GitHub with 3 skills. Review its manifest and source before installing.

by popoffvg

Work lifecycle management v2 — phased workflows (research, plan, implement) with agents, commands, and hooks.

by Mikita Taukachou

End-to-end issue workflows that orchestrate analysis, planning, implementation, verification, and shipping

by Fabian Distler

Planning and workflow skills: natural planning, repo status updates, release PRs for R and Python packages, and open-source contribution scouting.