enapt/SwarmLLM
Free, open-source app that runs AI chat models on your own computer — and lets computers team up over the internet to run models too big for one machine. No account, no crypto. OpenAI- and Anthropic-compatible API.
ARCHETYPE
Fort Knox
Deny lists, pre-tool hooks, sandboxing. Nothing touches prod without a signature.
GUARDRAILS
4/5
Blocks destructive commands · Protects secrets · Pre-tool screening hook · No YOLO mode · details
Copy this rig
# review before running: this installs third-party code
$ npx degit enapt/SwarmLLM/.claude ./rig-swarmllm # inspect, then merge into .claude/ MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit enapt/SwarmLLM/.claude/skills/check .claude/skills/check $ npx degit enapt/SwarmLLM/.claude/skills/cleanup .claude/skills/cleanup $ npx degit enapt/SwarmLLM/.claude/skills/next-task .claude/skills/next-task $ npx degit enapt/SwarmLLM/.claude/skills/review .claude/skills/review $ npx degit enapt/SwarmLLM/.claude/skills/sweep .claude/skills/sweep $ npx degit enapt/SwarmLLM/.claude/skills/test-module .claude/skills/test-module
$ curl -fsSL --create-dirs -o .claude/agents/root-cause.md https://raw.githubusercontent.com/enapt/SwarmLLM/main/.claude/agents/root-cause.md Merge into .claude/settings.json (project) or ~/.claude/settings.json (user). Hook commands reference scripts in the source repo — copy those too.
{
"permissions": {
"deny": [
"Bash(cargo publish *)",
"Read(./.env)",
"Read(./.env.*)",
"Read(./vendor/**/target/**/*)",
"Read(./target/**/deps/**)",
"Read(./target/**/build/**)",
"Read(./target/**/incremental/**)",
"Read(./target/**/.fingerprint/**)",
"Read(./target/**/examples/**)"
]
},
"hooks": {
"PreToolUse": [
{
"matcher": "Edit|Write|NotebookEdit|Bash",
"hooks": [
{
"type": "command",
"command": "bash ${CLAUDE_PROJECT_DIR}/.claude/scripts/research-gate.sh"
}
]
},
{
"matcher": "Bash",
"hooks": [
{
"type": "command",
"command": "bash ${CLAUDE_PROJECT_DIR}/.claude/scripts/commit-gate.sh"
}
]
}
]
}
} Skills (6)
Subagents (1)
| root-cause model: sonnet | Establishes whether a suspected cause actually causes an observed symptom, before any fix or revert is made. Use when about to attribute a failure to a change, a component, or a peer — particularly wh |
Hooks (4)
| event | matcher | runs |
|---|---|---|
| PreToolUse | Edit|Write|NotebookEdit|Bash | bash ${CLAUDE_PROJECT_DIR}/.claude/scripts/research-gate.sh |
| PreToolUse | Bash | bash ${CLAUDE_PROJECT_DIR}/.claude/scripts/commit-gate.sh |
| SessionStart | compact | bash ${CLAUDE_PROJECT_DIR}/.claude/scripts/post-compact-status.sh |
| InstructionsLoaded | * | bash ${CLAUDE_PROJECT_DIR}/.claude/scripts/instructions-loaded.sh |
Permissions
deny (9)
Bash(cargo publish *)
Read(./.env)
Read(./.env.*)
Read(./vendor/**/target/**/*)
Read(./target/**/deps/**)
Read(./target/**/build/**)
Read(./target/**/incremental/**)
Read(./target/**/.fingerprint/**)
Read(./target/**/examples/**)
ask (0)
—
allow (55)
Read
Edit
Write
Glob
Grep
Bash(cargo *)
Bash(cargo)
Bash(rustup *)
Bash(git *)
Bash(git)
Bash(ls *)
Bash(ls)
Bash(mkdir *)
Bash(wc *)
Bash(cat *)
Bash(head *)
Bash(tail *)
Bash(curl *)
Bash(find *)
Bash(grep *)
Bash(rustc *)
Bash(chmod *)
Bash(cp *)
Bash(mv *)
Bash(diff *)
Bash(echo *)
Bash(pgrep *)
Bash(ps *)
Bash(lsof *)
Bash(ss *)
Bash(tput *)
Bash(seq *)
Bash(test *)
Bash(expr *)
Bash(sort *)
Bash(uniq *)
Bash(cut *)
Bash(tr *)
Bash(xargs *)
Bash(stat *)
Bash(python3 *)
Bash(node *)
Bash(sed *)
Bash(awk *)
Bash(timeout *)
Bash(pip3 *)
Bash(sleep *)
Bash(docker *)
Agent(Explore)
Agent(Plan)
Agent(root-cause)
Agent(general-purpose)
Agent(feature-dev:code-explorer)
Agent(feature-dev:code-architect)
Agent(feature-dev:code-reviewer)
Similar rigs
mr-min-max/staledocs
Finds documentation that no longer matches your code. Deterministic AST check for pull requests; no API key. TypeScript, JavaScript, Python.
Pragmatist 2.8k tok ·
JuliusBrussee/cavekit
Frozen — compressed spec-driven development plugin for Claude Code. Still works; active development moved to JuliusBrussee/caveman.
Pragmatist 75 tok ·
warrendeanlangeveldt/code-kit
Keeps AI coding agents inside your project's rules: lanes, layer rules, protected paths and proof before finish, enforced by Claude Code hooks and in CI
Pragmatist 1.2k tok ·
gilles-g/localpr
Claude Code plugin: review a local git diff in a pull-request-style page, comment on lines, and let Claude apply the review. Python stdlib only, offline.
Pragmatist 5.4k tok ·