Find plugins

by chock-core

Best-effort guard against exfiltration through the tool channel: a network command (curl/wget/Invoke-WebRequest) that UPLOADS data -- POST/PUT, --data/--form, --upload-file, --post-file -- to a host outside the egress allowlist. The allowlist defaults to package registries and code hosting and is meant to be extended with your org's own domains; a host matches by exact name or ".<entry>" suffix. Fetch-only traffic is left alone. A schemeless or protocol-relative target is checked too, from the last non-flag token, but only when no explicit http(s):// URL appears -- once one has, it alone decides. Options are parsed as curl reads them, so -sd @file is an upload. Tool-time FLOOR, not a network sandbox. Known bypasses: ~/.curlrc, obfuscated payloads, non-standard clients, a language runtime. Escape: 'pragma: allowlist egress'. [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

Best-effort guard against exfiltration through the tool channel: a network command (curl/wget/Invoke-WebRequest) that UPLOADS data -- POST/PUT, --data/--form, --upload-file, --post-file -- to a host outside the egress allowlist. The allowlist defaults to package registries and code hosting and is meant to be extended with your org's own domains; a host matches by exact name or ".<entry>" suffix. Fetch-only traffic is left alone. A schemeless or protocol-relative target is checked too, from the last non-flag token, but only when no explicit http(s):// URL appears -- once one has, it alone decides. Options are parsed as curl reads them, so -sd @file is an upload. Tool-time FLOOR, not a network sandbox. Known bypasses: ~/.curlrc, obfuscated payloads, non-standard clients, a language runtime. Escape: 'pragma: allowlist egress'. [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

Best-effort guard against exfiltration through the tool channel: a network command (curl/wget/Invoke-WebRequest) that UPLOADS data -- POST/PUT, --data/--form, --upload-file, --post-file -- to a host outside the egress allowlist. The allowlist defaults to package registries and code hosting and is meant to be extended with your org's own domains; a host matches by exact name or ".<entry>" suffix. Fetch-only traffic is left alone. A schemeless or protocol-relative target is checked too, from the last non-flag token, but only when no explicit http(s):// URL appears -- once one has, it alone decides. Options are parsed as curl reads them, so -sd @file is an upload. Tool-time FLOOR, not a network sandbox. Known bypasses: ~/.curlrc, obfuscated payloads, non-standard clients, a language runtime. Escape: 'pragma: allowlist egress'.

by chock-core

Best-effort guard against an agent launching a coding agent with its safety checks off: claude --dangerously-skip-permissions or --permission-mode bypassPermissions, codex --full-auto, --yolo, --dangerously-bypass-approvals-and-sandbox or --sandbox danger-full-access, gemini --yolo, -y or --approval-mode yolo, cursor-agent --force. A spawned agent that never asks and never sandboxes is an unsupervised agent (OWASP ASI10, rogue agents). Read as a parsed command, so `cd repo && claude ...`, `bash -c '...'`, sudo/env wrappers and `npx @openai/codex ...` are caught, while a normal invocation, `codex --sandbox workspace-write`, or a command that only mentions the flag (echo, grep, a commit message) is not. Known bypass classes include shell aliases, wrapper scripts, config files that set the mode, and agents this list does not name. Run the agent with its default approvals; a human decides when an unattended run is acceptable. [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

Best-effort guard against an agent launching a coding agent with its safety checks off: claude --dangerously-skip-permissions or --permission-mode bypassPermissions, codex --full-auto, --yolo, --dangerously-bypass-approvals-and-sandbox or --sandbox danger-full-access, gemini --yolo, -y or --approval-mode yolo, cursor-agent --force. A spawned agent that never asks and never sandboxes is an unsupervised agent (OWASP ASI10, rogue agents). Read as a parsed command, so `cd repo && claude ...`, `bash -c '...'`, sudo/env wrappers and `npx @openai/codex ...` are caught, while a normal invocation, `codex --sandbox workspace-write`, or a command that only mentions the flag (echo, grep, a commit message) is not. Known bypass classes include shell aliases, wrapper scripts, config files that set the mode, and agents this list does not name. Run the agent with its default approvals; a human decides when an unattended run is acceptable. [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

Best-effort guard against an agent launching a coding agent with its safety checks off: claude --dangerously-skip-permissions or --permission-mode bypassPermissions, codex --full-auto, --yolo, --dangerously-bypass-approvals-and-sandbox or --sandbox danger-full-access, gemini --yolo, -y or --approval-mode yolo, cursor-agent --force. A spawned agent that never asks and never sandboxes is an unsupervised agent (OWASP ASI10, rogue agents). Read as a parsed command, so `cd repo && claude ...`, `bash -c '...'`, sudo/env wrappers and `npx @openai/codex ...` are caught, while a normal invocation, `codex --sandbox workspace-write`, or a command that only mentions the flag (echo, grep, a commit message) is not. Known bypass classes include shell aliases, wrapper scripts, config files that set the mode, and agents this list does not name. Run the agent with its default approvals; a human decides when an unattended run is acceptable.

by chock-core

The mechanizable slice of excessive agency, enforced at two points: at commit (the git hook, over staged changes) and at agent tool-use (over a tool call's arguments, as the agent writes) -- agent permission grants that allow everything. A settings file whose shell grant or allow-list is a bare wildcard hands the agent unlimited tool authority for every future session, in a file reviewers rarely read as code. The agent-world twin of block-wildcard-iam: scope grants to what the task needs (e.g. Bash(git status:*)). The allow/alwaysAllow/tools/defaultMode keys match whether or not YAML-style config quotes them, word-bounded so "disallow"/"allowlist" are not mistaken for "allow". Escape: 'pragma: allowlist broad-agency' on the same line. [Session-enforced at the turn's end by a Stop hook; needs git and a Python 3.11+. This client records no file-writing tool vocabulary, so the write itself is not judged: what the turn actually left on disk is re-read, and a construct a rule denies is refused then, however it was written. With no working Python the hook exits 2; without git, fail-open clients allow silently. A gate that cannot reach a decision refuses rather than allowing one it never judged. Enforcement at every commit and in CI still needs chock installed in the repo. Enforces only in a client that reads the com.github.copilot namespace and tells the hook where the package lives; a client that exports no plugin-root variable runs the hook, which then allows, so treat this package as advisory unless a deny has been witnessed in your own client.]

by chock-core

The mechanizable slice of excessive agency, enforced at two points: at commit (the git hook, over staged changes) and at agent tool-use (over a tool call's arguments, as the agent writes) -- agent permission grants that allow everything. A settings file whose shell grant or allow-list is a bare wildcard hands the agent unlimited tool authority for every future session, in a file reviewers rarely read as code. The agent-world twin of block-wildcard-iam: scope grants to what the task needs (e.g. Bash(git status:*)). The allow/alwaysAllow/tools/defaultMode keys match whether or not YAML-style config quotes them, word-bounded so "disallow"/"allowlist" are not mistaken for "allow". Escape: 'pragma: allowlist broad-agency' on the same line. [Session-enforced via PreToolUse and Stop hooks; needs git and a Python 3.11+. PreToolUse judges the file a tool call would write; Stop re-reads what the turn actually left on disk, so a file written through a shell heredoc is judged too. With no working Python the hook exits 2; without git, fail-open clients allow silently. A gate that cannot reach a decision refuses rather than allowing one it never judged. Enforcement at every commit and in CI still needs chock installed in the repo.]

by chock-core

The mechanizable slice of excessive agency, enforced at two points: at commit (the git hook, over staged changes) and at agent tool-use (over a tool call's arguments, as the agent writes) -- agent permission grants that allow everything. A settings file whose shell grant or allow-list is a bare wildcard hands the agent unlimited tool authority for every future session, in a file reviewers rarely read as code. The agent-world twin of block-wildcard-iam: scope grants to what the task needs (e.g. Bash(git status:*)). The allow/alwaysAllow/tools/defaultMode keys match whether or not YAML-style config quotes them, word-bounded so "disallow"/"allowlist" are not mistaken for "allow". Escape: 'pragma: allowlist broad-agency' on the same line.

by rickardp

Documented intent for AI-assisted development: ADRs, specs, patterns, requirements, DESIGN.md, and a separate design tree for UX decisions

by Paul Hinz, The Linux Foundation

LFX Marketing OS State of Foundation Board Update agent — produces the one-page board infographic (print-ready PDF plus HTML) for a Linux Foundation project or foundation ahead of each board meeting: key metrics, goals vs outcomes, risks, investment areas, and the decisions the board must take, tagged DISCUSS / REVIEW / VOTE. Reads the Quarterly Marketing Plan deck and the ED's board pre-read first, then interviews the ED for anything missing.

by Christopher Boone

Assess a repository, determine what scaffolding and setup tools are needed, present a plan, and execute them in the correct order.

by Christopher Boone

Assess a repository, determine what scaffolding and setup tools are needed, present a plan, and execute them in the correct order.

by Paul Hinz, Linux Foundation

Pulls the project's LFX record and runs a research sweep, then guides an LF project leader through a short intake (or pre-fills it in batch mode for ED review) and generates a project Brand Kit document — positioning, voice, audience messaging, competitive guardrails, LFX-derived governance and trademark wording, and a five-component visual identity direction — following the LFX Marketing OS Brand Kit standard — delivered as a Google Docs-ready document on the LF Agent DOCS Template, with the project's name and logo on the cover and its brand colors applied.

by Bugsee

Set up Bugsee SDKs, debug crashes and bug reports, inspect builds for size and dependency vulnerabilities, and upload symbols — via the Bugsee MCP server and skills.

by Mikita Taukachou

Conflict resolution, commit summaries, quick commits, and findings fixes

by duyet

A single skill that builds AI-agent applications end-to-end — interview or detect the stack, then scaffold UI/UX, agent APIs, tools, tracking and gateways across 13+ agent frameworks including LangGraph, the Vercel AI SDK, the OpenAI and Claude Agent SDKs, Google ADK, Mastra, Pydantic AI, and more — always verifying against live docs before writing code.

by keep-starknet-strange

Flagship Cairo/Starknet audit skill for vulnerability discovery with strict FP gating and confidence scoring.

by Eric Le

Automated governance, versioning, and discovery tools for Claude Code.

by Paul Hinz, Linux Foundation

LFX Marketing OS Case Study Agent — converts any source content (YouTube video, interview transcript, meeting notes, recording, or raw text) into an on-brand Word document for an LF project or foundation. Defaults to a case study; can also produce blog posts, articles, or summaries.

by Mingyang Bao

Academic research toolchain — systematic literature review, manuscript peer review, and IEEE reviewer discovery, backed by a single deterministic Python library and an accumulating local scholarly database.

by Mingyang Bao

LaTeX writing assistant — compile workflow, academic writing style, and word counting (texcount) for LaTeX documents

by fulin

資訊密度高的 Claude Code status line：目錄/repo/model/成本/用量/quota/agents/skills/memory+MCP/近期編輯/歷史多欄顯示，含 East-Asian 寬字對齊、原子寫入的跨進程狀態快取，與可用 /cc-statusline-rows 調整的顯示配置（逐列開關＋因機器而異的面板寬度邊距 widthMargin）。

by Jan Hoppe

cc-outline: the session as Markdown, git diffs, plans and a session browser in a split pane