Hooks plugins

Stop hook for Claude Code and Codex. When the agent ends with "all tests pass", "the build succeeds" or "lint is clean", it checks the session transcript for a matching command that ran after the last edit and exited 0, and blocks the stop if there isn't one. One file of standard-library Python, no LLM, no network.

1 skill · hooks

Stop/PostToolUse hook: after a code edit, Claude Code cannot end its turn until a recognised test command has run and passed after the last change. Piped, masked or failing runs do not count; after two reminders it lets the turn end with a warning. A repo's .proof-of-done.json can only add checks; waivers live in the user's ~/.claude/proof-of-done.json. Standard-library Python, no network.

1 skill · hooks

Automation Hooks - Event-driven automation hooks

Delegates work to a subagent or spawned session by writing only the task, while a PreToolUse hook pages the parent transcript through Jev and appends the chunks that task needs. Use before writing any Agent tool prompt or create_session prompt, and when asked to delegate this, hand this off, use a subagent, spawn an agent, run agents in parallel, or brief a subagent on what we have done so far.

1 skill · hooks

Load it without being asked at the start of any task whose deliverable is prose that another person will read, such as a PR description, commit message, README, doc, postmortem, blog post, issue or review comment, report, release note or essay, and run it on the draft before the text is pushed, posted, published or saved. Draft first without this skill in context, from a positive writing brief if one exists. Also use it when someone says "de-claude", "de-slop", "humanize this", "this reads like AI" or "make it sound human", or asks for a voice, tone or register edit. It rewrites the constructions that mark prose as model-written into plain technical prose, and checks that the rewrite kept the draft's claims. The constructions include staged reveals, verdict headers, aphoristic closers, "it's not X, it's Y", em-dash drama, forced triads, flat-certainty adverbs, noun stacks and dropped 'that'. Do not use it on fiction, poetry, code or quoted text. For a full adversarial review of a deliverable, use challenging.

1 skill · hooks

Automatically backup your work with git commits after each Claude session

hooks

Collect all TODO and FIXME comments from your project

hooks

Monitor code complexity and maintain code quality

hooks

Per-command credential guidance and best-effort cloud CLI stdout redaction. No session-wide credential loading. Powered by the nopeek CLI.

1 skill · hooks

Opens a question dialog when the context window passes a threshold (40% by default) and offers a handoff doc, without blocking background agents.

1 skill

Masks WordPress usernames and Application Passwords in prompts with session-scoped placeholders, and restores them into Bash commands at execution time.

AskUserQuestion 運用 hook 群（セッション単位無効化 #aq-off/#aq-on と発火時のデスクトップ通知）。

hooks

github.com への WebFetch を差し戻し gh コマンドへ誘導する hook。

hooks

Bash コマンドの sudo 実行を禁止する hook。

hooks

Auto-compile Elixir files with warnings-as-errors after edits

hooks

Auto-format Elixir files with mix format after edits

hooks

Run Credo static analysis after editing Elixir files

hooks

Blocks Claude Code from committing directly to main or master.

Runs lint and typecheck on staged files and blocks Claude's git commit if they fail.

Lints each file right after Claude edits it and feeds errors back to Claude.

Universal shell/git/tooling hygiene hooks - no personal or employer specifics

hooks

Pronoun-slip guard for nonbinary people and nonbinary plugin users

hooks

Morgan/Jasmeralia's personal-infra and workflow-policy guard hooks (TrueNAS, EC2, Odoo, memory policy, global CLAUDE.md tracking)

hooks

Universal shell/git/tooling hygiene hooks - no personal or employer specifics

hooks

Morgan/Jasmeralia's personal-infra and workflow-policy guard hooks (TrueNAS, EC2, Odoo, memory policy, global CLAUDE.md tracking)

hooks

Jira/Confluence/PagerDuty/Kibana workflow guard hooks

hooks

The lifecycle machinery the other plugins assume: shell-output compression that never rewrites a read whose bytes matter, a git guard, session context loading, quality gates, and structured logging. Wired across 18 events; every hook fails open, because a PreToolUse hook that exits non-zero blocks every shell call in the session.

hooks

Six safety hooks: block destructive shell commands, hardcoded secrets, permission-bypass flags and commits that do not type-check, check WebFetch URLs against an allowlist, and warn when a session starts in the home folder.

hooks

Block destructive shell and git commands before they run

Stop your agent from accessing unwanted or sensitive directories