Capsule

io.github.dawitlabsPublished by GitHub account dawitlabsv0.3.3Updated Oct 25, 2026

Compact, source-linked context packs that reduce token waste for coding agents.

VerifiedSTDIODesktop onlyKnowledge & MemoryDeveloper Tools

Overview

AI-generated overview

Capsule generates compact, source-linked Markdown context packs for a repository and serves them to coding agents over MCP to cut repeated discovery tokens.

What you can do
List capsules with freshness status and read a capsule's compressed context, with staleness warnings but no writes on read (capsule_list, capsule_get). Check which capsules have outdated sources and refresh one capsule or all of them (capsule_stale, capsule_refresh). Show token savings per capsule and machine-readable discovery-savings proof (capsule_stats, capsule_report). Run a CI-style stale, missing or config check, and generate capsules for the current repo (capsule_check, capsule_init). Expose capsules as MCP resources at capsule:///index and capsule:///{name}.
When to use it
Useful for large, multi-session coding work where agents keep rediscovering the same files, decisions, conventions and setup details, and you want a compact project context layer that any file-reading or MCP agent can consume.
Requirements
Runs locally over stdio as the npm package capsulectx, typically launched with npx capsulectx --mcp; Node.js is needed. Optional AI enrichment can use an installed Claude CLI, Claude or OpenAI API keys, or a generated prompt for a browser-based AI. No authentication or environment variables are declared in the manifest.
Before you install
Initialization and refresh write files into the repository, including a .capsules directory, and the tool can auto-patch CLAUDE.md, AGENTS.md, .cursorrules and .windsurfrules. Optional AI enrichment may send repository content to Claude or OpenAI or to a browser-based AI, so review what is shared. The project describes itself as early but functional.

Installation

In SourceWeft

  1. Open Capsule in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

Capsule

Compact, source-linked context packs for coding agents.

[capsule init and estimate demo]

[npm] [license] [node] [ci]

Docs · Quickstart · Commands

MCP gives agents tools. Skills give agents procedures. Capsules give agents compact project context.

Large repositories make coding agents burn tokens rediscovering the same files, decisions, conventions, and setup details over and over. Capsule targets a 50–70% reduction in repeated repository-discovery context for large, multi-session agent work.

Install

No install required — run directly with npx (always gets the latest version):

bash
npx capsulectx init

Or install globally with your package manager:

bash
# npmnpm install -g capsulectx
# pnpmpnpm add -g capsulectx
# yarnyarn global add capsulectx
# bunbun add -g capsulectx

Keep it up to date:

bash
npm install -g capsulectx --prefer-online

Tip: npx is the easiest option — it always runs the latest published version with no manual update step.

Quickstart

bash
npx capsulectx init

Capsule scans your repo, optionally enriches capsules with AI (using your installed Claude CLI, API keys, or a generated prompt for any browser-based AI), and creates:

.capsules/  index.md  architecture.md  setup.md  api.md  data.md  ui.md  testing.md  deployment.md

Each capsule is a Markdown file with source fingerprints in JSON frontmatter:

md
---json{  "name": "api",  "sources": ["src/api/**"],  "fingerprints": { "src/api/users.ts": "sha256:..." },  "updated_at": "2026-06-20T14:00:00.000Z"}---
# API Capsule
## Purpose
Request handlers, routes, controllers, and service boundaries.
## Key Files
- `src/api/users.ts`: source file matched by this capsule.
## Decisions
- Record stable decisions here so future agents do not rediscover them.

Agents read .capsules/index.md, choose the relevant capsule, check staleness, then inspect only the source files that actually matter.

Commands

CommandWhat it does
capsule initScan repo, run AI enrichment (optional), write all capsules
capsule enrichRe-run AI enrichment on existing capsules
capsule applyApply AI-generated JSON to capsule Conventions and Decisions
capsule write <name>Refresh one capsule from source (preserves your edits)
capsule refresh <name>Explicitly refresh one capsule
capsule refresh --allExplicitly refresh all detected capsules
capsule get <name>Print one capsule to stdout
capsule scanPrint detected source groups without writing
capsule stale [name]Check which source files changed since last write
capsule estimate <name>Show estimated token savings for one capsule
capsule statsShow token savings across all capsules
capsule report [name]Show reproducible source/capsule/stale token savings
capsule report --writeWrite .capsules/report.md as a proof artifact
capsule checkCI-friendly freshness/config check; exits nonzero on stale or missing capsules
capsule --mcpStart MCP server for native agent integration

MCP Server

Capsule ships an MCP server so Claude Code and other MCP-compatible agents can natively discover and consume capsules — no manual file reads needed.

Add to your project's .claude/settings.json:

json
{  "mcpServers": {    "capsule": {      "command": "npx",      "args": ["capsulectx", "--mcp"]    }  }}

Or globally in ~/.claude/settings.json to enable for all projects.

Once configured, agents get these tools automatically:

ToolWhat it does
capsule_listList all capsules with freshness status
capsule_getRead a capsule's compressed context; warns when stale but does not write
capsule_refreshExplicitly refresh one capsule or all capsules
capsule_staleCheck which capsules have outdated sources
capsule_statsShow token savings per capsule
capsule_reportShow machine-readable discovery-savings proof
capsule_checkCI-style stale/missing/config check
capsule_initGenerate capsules for the current repo

Capsules are also available as MCP resources at capsule:///index and capsule:///{name}.

Seeing the savings

$ capsule report architecture
Capsule: architecture
  capsule       status     source   capsule     stale   saved  architecture  fresh       9,607       417         0     96%

Token counts use the cl100k_base tokenizer for accurate measurement.

For CI, run:

bash
npx capsulectx check

It exits nonzero when capsules are stale, missing, or blocked by invalid config.

Agent Workflow

Add this to AGENTS.md, CLAUDE.md, Cursor rules, or your agent instructions:

md
Before working in this repo:
1. Read `.capsules/index.md` if it exists.2. Read the capsule matching the task area.3. Run `capsule stale <name>` when the CLI is available.4. If stale, inspect changed source files before editing.5. Update capsules when durable project knowledge changes.

Works with any agent that can read files — Claude Code, Codex, Cursor, Windsurf, Devin, custom MCP agents.

Custom Groups

Capsule detects standard layouts out of the box. For non-standard structure, add .capsules/config.json:

json
{  "groups": [    {      "name": "workers",      "description": "Background jobs, queues, and scheduled tasks.",      "sources": ["src/workers/**", "src/jobs/**"]    }  ],  "ignore": ["legacy/**"]}

A groups entry with the same name as a default group overrides it. New names are appended. ignore patterns extend the default ignore list.

See docs/capsule-format.md for the public capsule file format.

Building the Demo

The animation in this README is a generated GIF — no screen-recording tools required.

bash
npm run demo

Requires rsvg-convert (librsvg / librsvg2-tools) and ffmpeg. The script outputs docs/demo.gif from SVG frames rendered at 2× for retina. See scripts/build-demo.mjs.

Documentation Site

The OSS docs live in docs-site/ and are built with VitePress.

bash
npm run docs:devnpm run docs:build

Status

Capsule is early but functional. Current capabilities:

  • Language-agnostic repo scanning with monorepo support (Turborepo, pnpm workspaces)
  • Static analysis: framework detection, table/model extraction, route detection, env vars
  • AI enrichment on init — launches your installed Claude CLI, calls Claude/OpenAI API, or generates a prompt for any browser-based AI (Claude.ai, ChatGPT, Cursor, Windsurf)
  • Write-safe re-generation: human edits to Conventions and Decisions are preserved on refresh
  • MCP server for native Claude Code / agent integration
  • Real token counting (cl100k_base) with savings dashboard
  • TypeScript API surface extraction
  • Binary file filtering
  • Stale detection and token savings estimation
  • Reproducible report and CI-friendly check
  • Custom group config via .capsules/config.json
  • Auto-patches CLAUDE.md, AGENTS.md, .cursorrules, .windsurfrules with capsule instructions

Roadmap:

  • smarter write that diffs source changes and suggests targeted updates
  • before/after token reports for real agent sessions
  • GitHub Action wrapper for capsule check

Install for Local Development

bash
git clone https://github.com/dawitlabs/capsule.gitcd capsulenpm installnpm run buildnode dist/cli.js init

License

MIT

Source: README.md at commit 7a9aebf

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v0.3.3LatestSep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts