SuperSkill

Local project memory, rule-pack routing and worktree cache tools for coding agents

By Permanu TypeScript Local · stdio Knowledge and memoryDeveloper tools

40 popularity score ★ 19 ↓ 311/wk

Connect your client

Local (npm)

Add in one click

Opens the app, which asks before it adds the server.

Or add it by command or config

Claude CodeTerminal
claude mcp add superskill -- npx -y --prefer-online superskill@latest
Claude Desktopclaude_desktop_config.json
{
  "mcpServers": {
    "superskill": {
      "command": "npx",
      "args": [
        "-y",
        "--prefer-online",
        "superskill@latest"
      ]
    }
  }
}
Cursor~/.cursor/mcp.json
{
  "mcpServers": {
    "superskill": {
      "command": "npx",
      "args": [
        "-y",
        "--prefer-online",
        "superskill@latest"
      ]
    }
  }
}
VS Code.vscode/mcp.json
{
  "servers": {
    "superskill": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "--prefer-online",
        "superskill@latest"
      ]
    }
  }
}
Codex~/.codex/config.toml
[mcp_servers.superskill]
command = "npx"
args = ["-y", "--prefer-online", "superskill@latest"]
Gemini CLITerminal
gemini mcp add --scope user superskill npx -- -y --prefer-online superskill@latest

Settings 7

VAULT_PATH Optional: Folder of the markdown vault; must be under your home directory. Default ~/Vaults/ai.
MAX_INJECT_TOKENS Optional: Token cap for injected context, default 1500, clamped 100-50000.
SESSION_TTL_HOURS Optional: Hours before an agent session without heartbeat expires, default 2.
SUPERSKILL_MAX_RESPONSE_BYTES Optional: Byte budget per MCP response, default 65536, max 10 MiB.
SUPERSKILL_TELEMETRY Optional: 1 opts in to local-only telemetry, 0 turns recording off.
SUPERSKILL_CACHE_ROOT Optional: Root folder for shared worktree caches; defaults to the platform cache folder.
SUPERSKILL_FORCE_MCP Optional: Set to 1 to force server mode instead of the CLI.

What it does

SuperSkill is a local stdio MCP server, plus a CLI from the same npm package, that sits between a coding agent and the project it works on. Its main tool, `superskill`, takes a task description, picks curated rule packs by language, phase and specialty (for example Go, QA, security review), and returns a short brief and a rule plan cut to a token budget, with a record of what was picked and what was left out.

Around that it keeps a per-project memory: a markdown vault under `projects/<slug>/` (by default in `~/Vaults/ai`) holding decisions, tasks, learnings, sessions and notes, with an SQLite FTS5 index and link graph built from it. Notes can be opened in Obsidian, and `knowledge_viz` writes Mermaid architecture and code notes into the vault. Further tools cover spec, ticket and evidence gates, a tree-sitter code graph (`impact`, `claims`), sharing package and compiler caches across git worktrees, a read-only disk-space report, and a watchdog that reviews past agent sessions.

In version 0.15.1 the server registers 52 tools; the docs' tool table lists 51 and leaves out `skill_validate` (src/core/registry.ts). Calls are rate-limited per server in a 60-second window: 30 write calls, 60 reads and 10 expensive diagnostics, after which it returns `RATE_LIMITED` (src/mcp-server.ts).

What it needs

RuntimeNode.js 22+

Node.js 22.16 or later on the 22.x line, or Node 24+, built with SQLite FTS5; official Node 22.13 to 22.15 builds lack FTS5 and cannot run knowledge search. npx for the launch command. The vault folder (`VAULT_PATH`) must sit under your home directory. Running `superskill skill init` inside a repository is the docs' first step and makes vault tools find the project without naming it.

Limits

Vault access is confined to the current project's folder; cross-project search is refused, and a workspace that has not been registered cannot use implicit project memory. Responses over 64 KiB by default (`SUPERSKILL_MAX_RESPONSE_BYTES`, up to 10 MiB) come back as a truncated JSON preview. Context injection is capped at 1500 tokens by default (`MAX_INJECT_TOKENS`, 100 to 50000). The docs' launcher uses `superskill@latest` with `--prefer-online`, so each start fetches the newest release and may fail to start offline. The makers say the secret and prompt-injection scans are heuristics and that rule-pack effect on model output is unproven.

Who it suits

Suits a developer who runs coding agents (Claude Code, Cursor, OpenCode, Codex CLI) on their own machine and wants project decisions, tasks and session notes kept as local markdown, with rule packs chosen per task instead of a large skill catalogue. It also fits teams juggling many git worktrees who want shared caches and cleanup reports. It is a large tool surface (52 tools) that writes to your home directory and client configs, so someone who only wants a simple notes or memory store, or cannot run Node 22.16+, should pick something smaller.

Tools 52

superskill
Picks rule packs for a task and returns a brief plus a budgeted rule plan
read
Reads a vault file or lists a vault folder
write
Writes, appends or prepends to a vault file; appends by default
search
Full-text search over this project's vault notes
project_context
Returns the project's context document, found from the working directory
generate_context
Scans a repository and returns a draft context.md without saving it
decide
Records an architecture decision note
task
Adds, lists and updates project tasks or shows the board
learn
Saves or lists learnings
brainstorm
Starts or continues a brainstorm note
session
Registers, heartbeats, completes or lists agent sessions
resume
Shows recent sessions, unfinished work and next steps
stats
Counts the project's vault content
prune
Archives or deletes stale vault content by retention rules, dry run first
deprecate
Marks a vault item deprecated with a reason
init
Sets up SuperSkill for the current project
status
Shows knowledge-graph state: skills, weights, recent sessions
telemetry
Shows, enables, disables, reports or clears local usage telemetry
register
Maps a repository to a vault project
doctor
Runs a health check over install, vault, graph, catalog and clients
skill_install
Installs skills from a GitHub repository
skill_validate
Checks a native skill file against a profile without changing it
skill_list_installed
Lists installed skills
skill_remove
Removes an installed skill
link
Adds a wikilink between two vault notes
extract
Splits a source document into separate vault notes
capture
Saves several insights as separate vault notes
template
Gives prefilled templates for vault note types
graph_related
Finds notes related to one note in this project
graph_traverse
Walks project metadata and opens content in byte pages
graph_cross_project
Project-scoped search; other projects are refused
knowledge_rebuild
Rebuilds the search and link index from the markdown
knowledge_viz
Writes Mermaid architecture notes and code notes
snapshot_repo_state
Saves branch, changed files and last commit into the vault
env_facts
Stores or looks up stable facts about the environment
cred_refs
Stores pointers to where credentials are documented
rollback
Keeps rollback checkpoints for commits
spec
Creates, approves, freezes and lists plan specs
tickets
Creates and tracks tickets from frozen specs
evidence
Records or reads evidence for a ticket
gate
Checks whether a spec or ticket has the evidence it needs
impact
Finds definitions, importers and callers of code from a code graph
claims
Checks structured claims about code against a code-graph scan
worktree_status
Reports git worktree health and shared cache size
worktree_audit
Gives a safety verdict and cache duplication per worktree
worktree_env
Works out the shared-cache environment for this worktree
worktree_activate
Writes the repo cache policy and installs a post-checkout git hook
worktree_apply
Applies audit items or restores quarantined files after confirmation
worktree_gc
Reports old caches; quarantines or purges them after confirmation
worktree_uninstall
Removes the hooks and adapters it installed
hygiene_report
Reports disk space worth reclaiming, without deleting anything
watchdog
Reviews agent session traces and can apply approved fixes

Similar servers