Storybloq
Project memory and workflows for Claude Code and Codex: stories, plans, session handovers and review records kept in the repo
Medium risk
We rate an entry medium when the tool runs code, makes network calls or reads project files. Check what exactly it does before installing.
Why this level
- Reads and writes files in the repo
- Registers an MCP server and client hooks
- Can launch reviews and helper processes
Install
In your terminal, with SkillFoxx CLI
npx skillfoxx add cli/storybloqDetects the agents on your machine, checks the risk and pins the version.
Other ways to install
Assembled automatically, review before installing.
Install the tool
npm install -g @storybloq/storybloqInstall Storybloq: npm install -g @storybloq/storybloq@latest, then storybloq setup --client all. Call /story at session start to load project context.
Other ways from the author
npm install -g @storybloq/storybloq@latest
storybloq setup --client allsetup --client all installs the skill for Claude and Codex, registers the MCP server and configures hooks. Needs Node.js 20 and at least one client.
This is third-party code. Review the repository files before installing.
What it does
Storybloq keeps a .story/ directory of JSON and markdown in the repo: task stories, plans, roadmap phases, session handovers and captured lessons. These are plain files you can track with Git. The storybloq CLI reads and edits .story/ from the terminal. An MCP server gives Claude Code and Codex structured tools to read and update project state. A skill loads context at session start through /story in Claude Code or $story in Codex, and a separate free macOS app shows .story/ changes live.
Who it is for. For developers and leads who want decisions and unfinished work to survive across agent sessions.
Good fit when
- You want to continue in a new session without losing the previous one's decisions and context
- You want to keep task stories and handovers as files in the repo
- You need one source of project state for Claude Code and Codex
Not a fit when
- You need a full task tracker with a UI for the whole team
- The project does not use Claude Code or Codex
Example request
Load project state from .story/ and suggest which story to pick up nextLimitations
Requires Node.js 20 and at least one client: Claude Code or Codex CLI. The PolyForm Shield license restricts competing use and is not a classic open-source license. The macOS app is a separate product. Review and handover records help you inspect work but do not replace tests, CI and code review.
How to disable. Remove the global package: npm uninstall -g @storybloq/storybloq and unregister the MCP server in your client. The .story/ directory can be deleted from the repo manually.
Security check
- Reads and writes files in the repo
- Registers an MCP server and client hooks
- Can launch reviews and helper processes
README in short
The README frames the problem of context lost between agent sessions and proposes keeping project state next to the code. It installs globally via npm; storybloq setup --client all adds the skill, registers the MCP server and configures client hooks. It describes choosing the next story with blockers in mind, an autonomous mode with a plan and independent reviews, and handovers between sessions. It notes that approval and review records do not guarantee correct code. The package lives on npm and the macOS app ships separately.
FAQ
Where is data stored?
In the .story/ directory inside your repo, as JSON and markdown tracked by Git.
Do I need the macOS app?
No. It is a separate free product for watching .story/; the tool itself works through the CLI, MCP and skill.
Related
A skills library that gives coding agents a development process: brainstorming, planning, TDD, subagents and code review
Skills for real engineers by Matt Pocock
Skills For Real Engineers
Small composable skills for engineering with agents: plan grilling, TDD, bug diagnosis, code review and architecture
GitHub toolkit for spec-driven development: the specify CLI adds agent commands and skills to a project, from principles to implementation
Reference MCP servers
Model Context Protocol servers
Official reference MCP servers: Filesystem, Fetch, Git, Memory, Sequential Thinking, Time and Everything