claudecodex

thinking-tools

Metacognitive reasoning frameworks for careful decision-making. Includes Chesterton's Fence (investigate before removing), Pink Elephant (rewriting counterproductive negative instructions), Interview Mode (stop guessing after two rejected drafts and extract the user's own words), No Meat Proxy (rewrite a draft so the reader can relay it in their own words instead of forwarding it verbatim), Don't Lead the Witness (ask the way the real situation would, with no hint of the expected answer, so the response is genuine), and room for more thinking tools like pre-mortems, steelmanning, and first principles.

6 on GitHubby JT Sternbergv1.5.0MITupdated todayCoding
/

Plugin thinking-tools from marketplace jtsternberg in jtsternberg/claude-plugins. Commands read from the repo's manifest — today.

What it installs

Skills (5)

  • chestertons-fence

    Use before removing, refactoring, or 'simplifying' code, config, or processes whose purpose isn't fully understood — or when someone says 'Chesterton's Fence'. Understand why it was built before tearing it down.

  • dont-lead-the-witness

    Use when the user says 'don't lead the witness', or whenever you want someone's genuine, unprompted response: testing an agent or skill, asking a reviewer for a verdict, questioning the user, probing a real-world scenario. Ask the way the real situation would, with no hint of the answer you expect.

  • interview-mode

    Use when the user says 'interview me', 'interview mode', or 'stop guessing' — or when a second attempt at the same artifact (email, doc, plan, naming, UI copy, design) gets rejected or rewritten. Stop generating variants; interview the user and build from their own words.

  • no-meat-proxy

    Rewrite what you are about to hand the user so they can say it to a teammate in their own words, without opening a file. Manual invocation only — /thinking-tools:no-meat-proxy (Codex: $thinking-tools:no-meat-proxy) or "no meat proxy". Applies to the chat reply AND any draft headed to other humans (P

  • pink-elephant

    Use when the user flags a 'pink elephant' — an instruction phrased as a prohibition ('do not X', 'never X', 'avoid X') that draws attention to the very behavior it forbids. Rewrites negative rules into positive directives.

View source on GitHub jtsternberg.github.io

More from jtsternberg/claude-plugins

All 40 plugins →

hotline

claudecodex

Cross-workspace Claude Code communication. Dial another workspace to ask questions, delegate work, or collaborate in real-time.

9 skills

6

gws

claudecodex

Google Workspace CLI (gws) integrations — upload markdown to Google Docs, draft Gmail messages from markdown, manage Drive files, manage YouTube playlists, and more.

7 skills

6

agentmail

claudecodex

Give an agent its own email address. Send and receive email through the AgentMail API via the official agentmail CLI: create inboxes, send mail, read incoming messages and threads, reply / reply-all / forward, manage and schedule drafts, triage an inbox, keep a local address book of people and other agents, relay tagged work orders to another AI agent, and run the agent self-signup and OTP verification flow. Ships an opt-in hook that notices unread mail during a session. Triggers on mentions of AgentMail, agent email, inbox creation, sending or checking email for an agent, contacts or address lookup, inbox triage, and emailing another agent.

5 skills · hooks

6

session-tools

claudecodex

Tools for managing Claude Code session transcripts — read one out of a session, catch up on it, fork new work off it, generate recaps, or leave yourself a breadcrumb before pausing one of your parallel conversations. Shared transcript machinery lives at the plugin root so the skills cannot drift.

6 skills

6

skill-tools

claudecodex

Skills for creating and reviewing Claude Code extensions and validating shared skills under Claude Code and Codex.

6 skills

6

beads-workflow

claudecodex

Skills for working through beads epics and fixing findings with granular task and commit tracking.

4 skills · hooks

6

Similar plugins

mattpocock-skills

claude

Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).

39 skills

mattpocock/skills 323,108

superpowers

claudecodex

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

15 skills · hooks

obra/superpowers 316,836

claude-mem

claudecodex

Persistent memory for Claude Code: the next session starts where the last one left off.

29 skills · 1 MCP server · hooks

thedotmack/claude-mem 116,618

career-ops

claudecodex

Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications

5 skills · hooks

career-ops-hq/career-ops 82,253

mem0

claudecodex

Cross-session memory and token savings for coding agents.

6 skills · 1 agent · 1 MCP server · hooks

mem0ai/mem0 68,893

context7

claudecodex

Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.

1 MCP server

upstash/context7 64,709

Maintain thinking-tools?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex plugins40 plugins