agentrigs
+ submit rig
~ / rigs / ThinkFlowLab / system1-agents

ThinkFlowLab/system1-agents

System 1 decision models (Jev, Laya, Cua-S1) as brain for agents: Browser use, computer use, games and robotics

↗ GitHub ★ 119 apache-2.0 updated 3d ago collection Claude Code Claude plugin
share on X
ARCHETYPE
Pragmatist
A balanced, no-drama setup: some rules, some tools, nothing extreme.
CONTEXT TAX · EVERY TURN
~2.9k tokens
Moderate · median rig: 2.3k · breakdown
GUARDRAILS
0/5
No committed guardrails · details

Copy this rig

# review before running: this installs third-party code
$ claude mcp add s1a -- uv run --project ${CLAUDE_PLUGIN_ROOT:-.} s1a-mcp
$ npx degit ThinkFlowLab/system1-agents/.claude ./rig-system1-agents  # 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.

$ claude mcp add s1a -- uv run --project ${CLAUDE_PLUGIN_ROOT:-.} s1a-mcp
$ npx degit ThinkFlowLab/system1-agents/.agents/skills/add-agent-recipe .claude/skills/add-agent-recipe
$ npx degit ThinkFlowLab/system1-agents/.agents/skills/self-review .claude/skills/self-review
$ npx degit ThinkFlowLab/system1-agents/.claude/skills/build-s1a-agent .claude/skills/build-s1a-agent
$ npx degit ThinkFlowLab/system1-agents/.claude/skills/review-pr .claude/skills/review-pr
$ npx degit ThinkFlowLab/system1-agents/skills/s1a .claude/skills/s1a
$ curl -fsSL --create-dirs -o .claude/agents/s1a-browser.md https://raw.githubusercontent.com/ThinkFlowLab/system1-agents/main/agents/s1a-browser.md

This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 7,204 rigs:

{
  "permissions": {
    "deny": [
      "Read(./.env)",
      "Read(**/.env)",
      "Read(~/.ssh/**)",
      "Bash(rm -rf *)",
      "Read(**/*.pem)",
      "Bash(rm -rf /)",
      "Bash(git push --force:*)",
      "Bash(sudo *)",
      "Read(.env)",
      "Bash(rm -rf /*)",
      "Read(./.env.*)",
      "Read(~/.aws/**)",
      "Bash(git push --force*)",
      "Bash(rm -rf:*)",
      "Read(**/*.key)",
      "Read(**/.env.*)",
      "Bash(sudo:*)",
      "Bash(git reset --hard*)",
      "Bash(git reset --hard:*)",
      "Read(.env.*)"
    ],
    "ask": [
      "Bash(git push:*)",
      "Bash(git push *)",
      "Bash(git commit:*)",
      "Bash(rm *)",
      "Bash(rm:*)",
      "Bash(git rebase *)",
      "Bash(wget *)",
      "Bash(npm publish:*)",
      "Bash(git commit *)",
      "Bash(gh pr merge *)"
    ]
  }
}

MCP servers (1)

serversourceest. tokens
${claude_plugin_root:-.} · "s1a" pypi 2.5k

Skills (5)

add-agent-recipeself-reviewbuild-s1a-agentreview-prs1a

Subagents (1)

s1a-browser
model: inherit
Runs a web task with enumerable page controls through S1A (s1a), an openJiuwen browser agent with a System 1 decision model in its model slot, and returns the answer with where it was read. Use for cl

Similar rigs

tasnimAlam/dot-files
vim ( neovim ), zsh, doom emacs, kitty, fish shell and other configuration files.
★ 2 setup Claude Code 2 skills 2 hooks today
Pragmatist 110 tok ·
alondmnt/dotfiles
Claude Code configuration, skills, and dotfiles
★ 0 setup Claude Code 2 skills 2mo ago
Minimalist 1.2k tok ·
RapierCraftStudios/ForgeDock
ForgeDock — autonomous AI dev pipeline for Claude Code. GitHub as agent memory: issue in, PR out.
★ 117 project Codex 3 skills today
Minimalist 724 tok ·
eycjur/dotfiles
My dotfiles
★ 0 setup Claude CodeCodex 3 skills 1d ago
Minimalist 510 tok ·
copied ✓