By Eigenwise
A Trello-light quest log for Claude Code. The side issues you mention mid-task ("oh, and the contact form doesn't send") get captured as tickets on the spot, with any pasted images attached, and managed on a live, self-hosted Kanban dashboard that spans every project you work in.
Open the sidequest board (live Kanban of your tickets) in the browser
Run a full board-grooming pass over a sidequest project, like a sprint grooming session: sweep every ticket, cross-check it against reality (git log, docs, recent session work) to find done-but-open tickets, superseded ones, stale claims, duplicates, and missing tickets for work that already exists in the repo, then act on what's clearly safe and batch the unclear ones into a few interactive questions for the user before closing anything ambiguous. Use when the user says "groom the board", "board grooming", "sprint grooming", "clean up the board", "tidy tickets", "audit the tickets", or "is the board still accurate". Never deletes tickets (closes with an evidence-bearing comment instead) and never touches a claim held by an active agent.
Open the sidequest board (a live Kanban of tickets) or manage tickets from the CLI/MCP: file, list, update, move, close, prioritize, label, or delete tickets — "show me the dashboard", "close SQ-3". Use to WORK the board ("grab the next task") — atomically CLAIM first. Use when the user hands you substantial or multi-part work — decompose it into linked tickets BEFORE implementing. Use to comment, ask the user on-ticket questions (a question means pause-and-wait), or relate tickets. Categories drive model/effort routing. For a mid-task side issue, file it with `add` and keep working. Filing never asks you to work it.
Matches all tools
Hooks run on every tool call, not just specific ones
Admin access level
Server config contains admin-level keywords
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Executables (bin/) — files in this plugin's bin/directory are added to the Bash tool's PATH while the plugin is enabled.
Executes bash commands
Hook triggers when Bash tool is used
Executes bash commands
Hook triggers when Bash tool is used
A small, growing marketplace of Claude Code plugins by Eigenwise.
Sharp little tools for Claude Code, kept in one shed. 🛠️
| Plugin | What it does |
|---|---|
| workbench | The one user-scope install for workspace setup, Toolshed updates, health checks, retrospectives, and the freshness guard. Use /init-workspace, /update-toolshed, /workbench-doctor, and /retro. |
| codebase-mapper | Keeps a small, self-updating map of your codebase and loads it into every Claude session, so Claude already knows how your project is built when you start working. |
| live-rules | Inject your own rules into Claude's context the moment they apply: global rules on every prompt, file-type and directory rules right before an edit, keyword rules when your prompt matches. Edit a rule, it applies on the next prompt. |
| sidequest | A Trello-light quest log for Claude Code. Side issues you mention mid-task get captured as tickets on the spot, with pasted images attached, then managed on a live, self-hosted Kanban dashboard spanning every project you work in. Category-based routing picks a concrete model and reasoning effort per ticket, with category and global fallbacks. |
More tools will move into the shed over time.
/plugin marketplace add Eigenwise/eigenwise-toolshed
/plugin install workbench@eigenwise-toolshed --scope user
/plugin install codebase-mapper@eigenwise-toolshed
/plugin install live-rules@eigenwise-toolshed
/plugin install sidequest@eigenwise-toolshed
Then run /reload-plugins (or restart Claude Code) and you're set. It's a public marketplace, so there's no auth to deal with.
Quickest start: install workbench at user scope, reload, then in any project ask "set up a Claude workspace here". Workbench owns setup, updates, health, retrospectives, and the freshness guard.
Workbench owns the workspace lifecycle in one user-scope install. It sets up project-side configuration, updates Toolshed plugins, checks local health, runs retrospectives, and covers freshness checks.
On any non-trivial repo, a fresh Claude session starts blind. It greps for the architecture, hunts down the entry points, and re-learns your conventions before it can do anything useful, then starts from scratch again the next session. codebase-mapper does that work once and keeps it:
.claude/.codebase-info/ (architecture, modules, entry points, patterns, dependencies, and so on).UserPromptSubmit hook re-injects the map on every prompt, so it stays in context deep into a long session and Claude keeps consulting and updating it as you work.Two skills run it: map-codebase builds the map and update-codebase-map refreshes it. The plugin README has the full details.
CLAUDE.md is a static, always-on brief. But a lot of guidance is conditional: a React rule only
matters when you touch a .tsx file, a deploy checklist only matters when you deploy. Put it all in
CLAUDE.md and it is either permanently in your context or quietly buried. live-rules fixes that:
npx claudepluginhub eigenwise/eigenwise-toolshed --plugin sidequestSkills plus explorer and reviewer subagents for building, scaffolding, debugging, understanding, and auditing applications with the Atomic Agents Python framework. Includes the umbrella `framework` skill, action-oriented `create-atomic-schema` / `create-atomic-agent` / `create-atomic-tool` / `create-atomic-context-provider` skills, the `troubleshoot` diagnostic skill, the `new-app` scaffolder (emits AGENTS.md + CLAUDE.md), progressive-disclosure reference material for prompts, orchestration, memory, hooks, providers, project structure, and testing, and isolated-context subagents for codebase mapping and code review.
MCP gateway that lets any web app expose typed actions to Claude over WebSocket — no browser automation. Reverse-connection architecture: the gateway is a pure WebSocket client that discovers apps via `~/.tesseron/tabs/` — no fixed ports, no env vars. Plus a `framework` skill for the Tesseron mental model, a `tesseron-docs` skill that calls the published `@tesseron/docs-mcp` server for authoritative chapter-and-verse spec lookups, a `tesseron-dev` skill that picks the right consumer package (`@tesseron/react` for React, `@tesseron/svelte` for Svelte 5, `@tesseron/vue` for Vue 3, `/server` for Node, `/web` for any other browser context) and inserts the canonical API at the project's entry point, and `tesseron-explorer` / `tesseron-reviewer` skills for mapping and auditing Tesseron codebases.
Project rules in a Markdown file that Claude re-injects right when they apply: every prompt, or just before it edits a matching file. Edits apply on the next prompt, no restart. A SessionStart hook injects the always-on rules once as a fallback, and it's a way to tell whether per-prompt injection is actually wired.
Set up and maintain a Claude Code workspace: initialize project-side configuration, check and update Toolshed plugins, diagnose local health, and run retrospectives.
A self-maintaining codebase map: atomic docs Claude loads at the start of each session and keeps in sync as the code changes. Any stack; greenfield or existing.
Live project kanban for Claude Code. Auto-tracks shipped / fixed / deferred work as cards on a board.json + live HTML board (127.0.0.1:7891), survives sessions via a SessionStart digest, auto-links edited files to their cards, and backstops un-carded work on Stop. Bootstraps a fresh board by mining past session history.
Personal task tracker — invisible kanban that tracks your work automatically
Cotask — task management with TASKS.md kanban dashboard
Plan a huge chunk of work — more than one agent session can hold — as a shared map of investigation tickets on your issue tracker, and resolve them one at a time until the way to the destination is clear.
Manage long-running development projects across multiple sessions
Linear issue tracking integration and workflow management