claudecodex

workbench

The complete Workshop toolkit — every skill, agent, methodology doc, and safety hook in one package, including the vault lifecycle, graph, capture, search, sync, and writing workflows. Skills and agents install on Claude Code, Codex, and Cortex Code; the safety hooks execute on Claude Code and Cortex Code. Plan, build, and ship with the full first-party dev workflow.

0 on GitHubv8.47.5updated 19 days agoProductivity
/

Plugin workbench from marketplace the-workshop in cdcoonce/the-workshop. Commands read from the repo's manifest — 19 days ago.

This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.

What it installs

Skills (81)

  • adversarial-review

    Attacks finished work by trying to disprove what it claims, reporting what survives with evidence. Use when the user says "attack this" or "try to break this", or wants a hostile pass before shipping, merging, or trusting a claimed-done result.

  • blueprint

    Charts an effort too big for one session as a map of decision tickets, worked one ticket at a time. Use when invoking /blueprint, saying "chart a map" or "work the map", or planning work too large to spec at once.

  • brainstorm

    Interviews the user like a business analyst to shape a fuzzy idea into a committed direction before any plan, PRD, or code exists. Use when the user wants to brainstorm, explore approaches to a problem, or has a goal but no committed solution yet.

  • build-review-land

    Ships one PR via a builder, a blind mutation reviewer who commits a re-runnable teeth spec, a RED-confirm fixer, and a pinned landing. Use for "worker-per-PR + mutation review", "build this PR with a worker and a reviewer", or the skywatch recipe.

  • chart-taste

    Applies chart-design taste to React data viz — a chart-type decision tree and dials for density, complexity, color restraint. Use when building charts with Recharts, Nivo, or similar libraries. Not container chrome — use react-ui-ux.

  • codex-enable

    Vendors a Codex-functional .codex/hooks.json into a target repo so a Codex teammate gets workbench's hook protections. Use when someone asks to enable workbench hooks on Codex or to set up .codex/hooks.json in a repo.

  • commit

    Stages and commits changes using enforced conventional commit style with atomic, well-scoped commits. Use when Claude needs to craft a commit message or the user asks to commit, make a commit, or save their work.

  • create-hook

    Create and register Claude Code hooks (PreToolUse, PostToolUse) as Python scripts. Use when user wants to create a hook, add a pre-edit check, post-edit formatter, block file edits, or automate responses to tool use.

  • daa-code-review

    AI-powered code quality review for Python, Markdown, and Mermaid. Use when the user asks for a "code review" or "quality check", or pastes code for analysis. Not vulnerabilities — see security-review.

  • dagster-expert

    Expert guidance for Dagster and the dg CLI — assets, components, schedules, sensors, jobs. Use before tasks referencing Dagster, materialization, or data pipelines, including creating a project or finding definitions.

  • data-discovery

    Generates a data discovery doc for a Snowflake schema or dbt project: table inventory, coverage, ER diagram, and runnable queries. Use when onboarding to a dataset or asking what data exists. Not dbt structure — use dbt-manifest-facts.

  • dbt-expert

    Expert guidance for building dbt Core models, sources, tests, and snapshots. Use before tasks referencing dbt, ref()/source(), materializations, YAML schema, or dbt build/run/test. Not project auditing — use dbt-manifest-facts.

  • dbt-manifest-facts

    Answers structural questions about a dbt project from its manifest.json, not comments or memory. Use when counting models/tests/seeds, auditing a repo, or tracing what feeds a model. Not dbt-expert or data-discovery.

  • design-an-interface

    Generate multiple radically different interface designs for a module using parallel sub-agents. Use when user wants to design an API, explore interface options, compare module shapes, or mentions "design it twice".

  • detector-teeth-check

    Prove tests catch the defect they claim to prevent: re-inject it, check the suite goes red. Use for any teeth check, mutation run or vacuous-test doubt, and in place of hand-rolling a mutate/revert script.

  • dev-cycle

    Takes a GitHub-issues-driven feature from brainstorm through a merged PR, interactively. Use when the user says "dev cycle", "development workflow", or "full development pipeline", or invokes /dev-cycle.

  • dignified-python

    Production Python coding standards (3.10-3.13): modern type syntax, LBYL exception handling, pathlib, ABC-based interfaces. Use when writing, reviewing, or refactoring Python in any project, not only Dagster.

  • drain-queue

    Works a queue of specced issues by hand, one worker per issue, with a spec gate and review before landing. Use when ready issues must ship as a batch, or the user says "work the queue", "drain the backlog", or "build these by hand".

  • drill-me

    Closed-book multiple-choice drilling that tells a recall gap apart from a comprehension gap. Use when Charles invokes /drill-me or asks to be drilled, quizzed, or tested on material — especially before an assessment.

  • explain

    Explain something briefly in chat (eli5) or build a one-page visual HTML summary of a session, branch, work tree, or MR. Use for eli5/plain-english asks and summary, recap, or status page asks. Not vault wrap-up/handoff and not transcript-notes.

  • finish-branch

    Use when implementation is complete, all tests pass, and you need to decide how to integrate a finished development branch — merge, open a PR, keep it, or discard it. Do not use mid-implementation or while the test suite is red.

  • github-cli

    GitHub CLI (gh) integration for issues, pull requests, branches, commits, and Actions workflows from the terminal. Use to create, list, or update GitHub issues, open or review pull requests, or manage Actions without a browser.

  • github-pr-land

    Lands a GitHub PR only at a CI-tested head: refreshes a moved base, re-gates, waits on that exact SHA, merges pinned. Use to merge, land, or ff-promote a PR, or to watch CI on a pushed SHA.

  • gitlab-ci-watch

    Watches GitLab CI until a pushed commit, MR, or integration branch reaches a terminal state. Use after pushing to GitLab, on a `glab mr merge` 405, or to confirm post-merge CI on dev. For browsing, use gitlab-cli.

  • gitlab-cli

    GitLab CLI (glab) for issues, MR review, and CI/CD pipelines. Use to create, list, or update GitLab issues, review or merge MRs, or inspect pipelines and logs. For MR creation use gitlab-mr-create; for CI watching use gitlab-ci-watch.

  • gitlab-mr-create

    Create GitLab merge requests with `glab` — into `dev` a conventional-commit title (`HEAD` subject or a title file), into `staging` and `main` a title file; descriptions keep real newlines; both read back. Use whenever creating a GitLab merge request.

  • gitlab-promotion-flow

    Integration and promotion policy for Clearway GitLab data repos (Dagster, dbt, ingestion). Use when starting work, choosing a branch or merge target, opening or merging an MR into dev, promoting dev to main, or releasing to production there.

  • grill-me

    Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".

  • hold-scan

    Checks whether anyone recorded that a PR or MR must not land: thread text, afk:hold labels, and an optional notes vault. Use before merging any PR or MR, or before trusting a "ready to merge" brief. Read-only.

  • live-view-simulate-diff

    Simulates a proposed change to a live Snowflake view read-only and diffs it against the deployed output with float tolerance. Use when a view has no committed source (co-edited in the UI) and must be verified before a role-gated manual deploy.

  • measure-before-landing

    Run a two-arm comparison that leaves a committed ledger: prereg enforced, fixture fingerprinted, arms bracketed A-B-A, exact sign test and MDE reported. Use when you A/B, benchmark, baseline, or ask whether a change actually improved anything.

  • mr-merge-order

    Computes a pairwise conflict matrix with git merge-tree and recommends a merge order by rebase cost. Use when several MRs or PRs target the same branch and the user asks which to merge first or how to order a queue. Read-only.

  • mr-preflight

    Catches stale references before a GitLab MR opens: names a diff renamed that code, SQL or docs still use. Use when creating an MR into dev, when create-mr refuses with an mr-preflight hit, or to waive or permanently ignore an old name.

  • mr-review-fixes

    Reads MR/PR review threads, triages findings, lands the fix, and replies to the reviewer in-thread. Use when an MR or PR has review feedback, changes requested, or an approval blocker, or asks what needs fixing after review.

  • persona-builder

    Builds an installable, self-tuning coach/advisor persona for a named owner via interviews, field research, and a three-layer package delivered by PR. Use when asked to create a persona, coach, sounding board, or advisor for someone.

  • plan-ceo-review

    CEO/founder-mode review that rethinks a plan to find the 10-star product. Use when the user asks for a plan review, CEO review, mega review, or wants a plan challenged or stress-tested before implementation.

  • prd-to-issues

    Breaks a PRD into tracer-bullet vertical slices as executor-ready GitHub issues, or a phased plan under docs/plans/ with `--plan`. Use to convert a PRD to issues, or break a PRD into work items or tickets, or mentions "tracer bullets".

  • project-context

    Generates or updates `.claude/docs/project.md`, the dense project reference Claude loads every session. Use when asked to create, update, or regenerate project context, or says "update project.md". Not the README — see repo-docs.

  • react-ui-ux

    Applies design taste to React UI generation using variance/motion/density dials, stopping components from the generic shadcn/Tailwind look. Use when building or editing React components/pages. Not charts — use chart-taste.

  • repo-docs

    Writes and maintains a repo's README and docs/ (tutorials, how-to, reference, explanation). Use when asked to write or refresh a README, runbook, or architecture doc, explain where X lives, or check docs for staleness. Not project.md or docs/plans.

  • request-refactor-plan

    Use when user wants to plan a refactor, create a refactoring RFC, break a refactor into safe incremental steps, or find architectural improvement opportunities (deepening shallow modules, consolidating tightly-coupled code, making a codebase more testable or AI-navigable). Produces a detailed refact

  • security-review

    Security code review for vulnerabilities with confidence-based reporting. Use when the user asks for "security review", "find vulnerabilities", or an OWASP audit. Not general code quality — see daa-code-review.

  • session-orchestrator

    Supervises user-visible worker sessions across projects and platforms. Use when a controller must create, coordinate, monitor, redirect, integrate, or retire multiple Codex, Claude Code, or Cortex Code worker tasks.

  • shared-tree-safety

    Check who else uses a git tree, index, venv or cache before you reset, rebuild, delete, prune or call it dormant: live processes, MCP and launchd registrations, other sessions. Use before `reset --hard`, `clean`, `reindex --force` or rm.

  • sql-deploy-precheck

    Compile-checks warehouse SQL (Snowflake, BigQuery, Redshift) against the live schema before deploying, catching drift. Use when about to run a .sql file or after an ALTER TABLE. Not fixtures — use warehouse-sql-test-harness.

  • stale-artifact-sweep

    Re-verifies a recorded issue, review finding, "do not merge" comment, TODO, or blocker against current reality, with evidence. Use before acting on an old claim that something is blocked, broken, or still needed. Read-only.

  • tdd

    Test-driven development with red-green-refactor loop. Use when user wants to build features or fix bugs using TDD, mentions "red-green-refactor", wants integration tests, or asks for test-first development.

  • transcript-notes

    Turns a YouTube lecture or transcript (VTT, SRT, text) into a structured Obsidian study note with reconstructed math and flagged visuals. Use when given a YouTube URL or a transcript to make readable. Not summarizing or quizzing.

  • triage-issue

    Use when user reports a bug, wants to file an issue, mentions "triage", or wants to investigate and plan a fix for a problem. Finds the root cause and files a GitHub issue with a TDD-based fix plan.

  • triage-quarantine

    Diagnose and resolve a failed, quarantined, or question-parked autonomous agent run, reusing its preserved work instead of rebuilding. Use when a slice quarantines, exits nonzero, or gets rejected by a gate.

  • using-workflow

    Use when starting any conversation or task in this project — establishes precedence between instructions and skills, requires invoking any skill that might apply, and sets the order skills run in before any response or action.

  • vault-audit

    Run Charles's vault (The Vault) /vault-audit structural audit across frontmatter, wikilinks, indexes, stale notes, duplicates, and templates. Trigger when Charles invokes /vault-audit, mentions /vault-audit, or asks for this vault workflow by name.

  • vault-budget

    Run Charles's vault (The Vault) /budget spend and subscription-value meter from local Claude transcripts. Trigger when Charles invokes /budget, mentions /budget, or asks for this vault workflow by name.

  • vault-clickup-task-sync

    Run Charles's vault (The Vault) /clickup-task-sync workflow to sync vault action items into ClickUp without duplicating tasks. Trigger when Charles invokes /clickup-task-sync, mentions /clickup-task-sync, or asks to sync action items / a 1:1 recap into ClickUp.

  • vault-cold-read

    Run Charles's vault /cold-read gate — an adversarial read of an issue's SPEC before promotion to the afk executor. Trigger when Charles invokes /cold-read, or asks if an issue is buildable cold.

  • vault-connect

    Run Charles's vault (The Vault) /connect autonomous graph connection pass with preview-gated wikilink edits, including a manual `--term` mode for finding or inserting the right wikilink for a note. Trigger when Charles invokes /connect, mentions /connect or --term, asks to find the right wikilink fo

  • vault-context-then-delegate

    Runs Charles's vault /context-then-delegate workflow, resolving email/SharePoint/Slack ambiguity before writing a prompt. Trigger: Charles invokes /context-then-delegate, or is about to write a delegated prompt.

  • vault-debrief

    Run Charles's vault (The Vault) /debrief retrospective over recent afk builds. Trigger when Charles invokes /debrief, mentions /debrief, asks how much of the autonomous work needed his hands, asks whether /cold-read is actually earning its cost, or asks which stage keeps producing rework.

  • vault-dispatch

    Run Charles's vault (The Vault) /dispatch workflow to turn a shaped idea into an afk-managed issue linked back into the vault. Trigger when Charles invokes /dispatch, mentions /dispatch, or asks for this vault workflow by name.

  • vault-dump

    Run Charles's vault (The Vault) /dump capture workflow for routing freeform input into durable vault notes, tasks, indexes, and wikilinks. Trigger when Charles invokes /dump, mentions /dump, or asks for this vault workflow by name.

  • vault-essay

    Draft long-form prose (essays and posts) in Charles's voice using The Vault's /essay writing rules. Trigger when Charles invokes /essay, mentions /essay, or asks to draft an essay or post in his voice.

  • vault-find

    Run Charles's vault (The Vault) /find semantic vault search workflow, including reindex and status modes. Trigger when Charles invokes /find, mentions /find, or asks for this vault workflow by name.

  • vault-fix-issue

    Run Charles's vault (The Vault) /fix-issue workflow to resolve a filed issue under TDD + mutation-teeth-check + review-before-commit discipline. Trigger when Charles invokes /fix-issue, mentions /fix-issue, or asks to fix/resolve a specific filed issue with test rigor.

  • vault-garden

    Run Charles's vault (The Vault) /garden graph-gardener apply workflow for queued link, profile, memory, index, and orphan repairs. Trigger when Charles invokes /garden, mentions /garden, or asks for this vault workflow by name.

  • vault-grill

    Run Charles's vault (The Vault) /grill active knowledge-extraction interview and route the result into the vault graph. Trigger when Charles invokes /grill, mentions /grill, or asks for this vault workflow by name.

  • vault-handoff

    Run Charles's vault (The Vault) /handoff workflow to refresh the machine-scoped rolling handoff digest. Trigger when Charles invokes /handoff, mentions /handoff, or asks for this vault workflow by name.

  • vault-init

    Run Charles's vault (The Vault) /vault-init workflow to scaffold a brand-new second-brain vault from the-workshop's vault-ops machinery. Trigger when Charles invokes /vault-init, mentions /vault-init, or asks to create, bootstrap, or stand up a new vault for himself or someone else.

  • vault-mr-review-packet
  • vault-podcast
  • vault-pulse
  • vault-recall
  • vault-standup
  • vault-sync
  • vault-teach
  • vault-wrap-up
  • vault-write
  • walkthrough
  • warehouse-sql-test-harness
  • worktree-audit
  • write-a-prd
  • xlsx-template-row-edit

Hooks (19)

  • ConfigChange

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-hook.sh audit-config-change.py

  • PostToolUse

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-hook.sh post-edit-lint.py

  • PostToolUse

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-vault-hook.sh vault-validate-write.py

  • PostToolUse

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-hook.sh remind-skill-announce.py

  • PreCompact

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-vault-hook.sh vault-pre-compact.py

  • PreToolUse

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-hook.sh branch-policy.py

  • PreToolUse

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-hook.sh protect-files.py

  • PreToolUse

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-hook.sh refuse-held-merge.py

  • SessionEnd

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-hook.sh warn-off-trunk.py

  • SessionStart

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-hook.sh inject-skill-router.py

  • SessionStart

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-vault-hook.sh vault-session-start.py

  • Stop

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-vault-hook.sh vault-stop-1-notebook-update.py

  • Stop

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-vault-hook.sh vault-stop-2-graph-gardener.py

  • Stop

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-vault-hook.sh vault-stop-3-session-sync.py

  • Stop

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-hook.sh verify-tests-before-stop.py

  • SubagentStart

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-hook.sh snapshot-subagent-start.py

  • SubagentStop

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-hook.sh verify-subagent-evidence.py

  • UserPromptSubmit

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-hook.sh suggest-handoff-on-context.py

  • UserPromptSubmit

    bash "${CLAUDE_PLUGIN_ROOT:-${PLUGIN_ROOT:-$WORKSHOP_PLUGIN_ROOT}}"/hooks/run-vault-hook.sh vault-skill-alias.py

View source on GitHub

More from cdcoonce/the-workshop

All 9 plugins →

Tools for auditing and maintaining The Workshop's skills, plugins, and distribution boundaries

4 skills

0

Product-design/UI-UX advisor persona — artifact-first design reviews with severity-tagged findings, named principles, and a stance contract that holds positions against pushback. Built by persona-builder.

1 skill

0

Product-strategy sounding board and coach persona for a design+PM hybrid at an early-stage startup — decision stress-testing with a steelman duty, influence-case building, prioritization on thin evidence, and verdict-first design critique. Built by persona-builder.

1 skill

0

Collaborative pair-programmer voice — brief think-aloud, checks in at decision points.

hooks

0

persona-ship-it

claudecodex

Momentum-first voice — blunt, bias-to-action, picks a sensible default and moves.

hooks

0

Senior-staff-engineer voice at full depth — reasoning, tradeoffs, and edge cases spelled out.

hooks

0

Similar plugins

caveman

claude

Talk like caveman. Cut filler. Keep technical details explicit.

22 skills · 1 command · 3 agents · hooks

JuliusBrussee/caveman 119,462

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

Taste Skill

claudecodex

tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.

1 skill

Leonxlnx/taste-skill 114,391

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

pake

claudecodex

Package any website or local web build into a lightweight desktop app with Pake CLI.

1 skill

tw93/Pake 63,463

Maintain workbench?

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 plugins9 plugins