uncoalesced/ArgusCouncil
Provenance locked, three agent research harness for any AI coding agent. A Scout downloads sources, an Archivist hashes and quotes them, an Integrator cites them into your paper, report, or dataset. A stdlib Python engine verifies every quo
ARCHETYPE
Minimalist
Lean instructions, few tools, tiny context tax. Lets the model think.
Copy this rig
# review before running: this installs third-party code
$ npx degit uncoalesced/ArgusCouncil/.claude ./rig-arguscouncil # 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 uncoalesced/ArgusCouncil/.agents/skills/council .claude/skills/council $ curl -fsSL --create-dirs -o .claude/agents/archivist.md https://raw.githubusercontent.com/uncoalesced/ArgusCouncil/main/.claude/agents/archivist.md $ curl -fsSL --create-dirs -o .claude/agents/integrator.md https://raw.githubusercontent.com/uncoalesced/ArgusCouncil/main/.claude/agents/integrator.md $ curl -fsSL --create-dirs -o .claude/agents/scout.md https://raw.githubusercontent.com/uncoalesced/ArgusCouncil/main/.claude/agents/scout.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 6,974 rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(rm -rf *)",
"Read(**/*.pem)",
"Bash(rm -rf /)",
"Bash(git push --force:*)",
"Bash(sudo *)",
"Read(~/.aws/**)",
"Bash(rm -rf /*)",
"Read(./.env.*)",
"Read(.env)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Read(**/.env.*)",
"Read(**/*.key)",
"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(npm publish:*)",
"Bash(wget *)",
"Bash(git rebase *)",
"Bash(gh pr merge *)",
"Bash(git commit *)"
]
}
} Skills (1)
Subagents (3)
| archivist model: inherit | Argus Council Archivist. Verifies each downloaded source and completes its entry in research/LEDGER.md (metadata, APA citation, reliability, verbatim quoted facts). Use for council phase P2, REQUESTs |
| integrator model: inherit | Argus Council Integrator. Connects verified ledger evidence to the task and writes it into the deliverable named in research/PLAN.md with [S###.F#] citations. Use for council phase P3, fixes to the de |
| scout model: inherit | Argus Council Scout. Searches the web and downloads evidence for the questions in research/PLAN.md, registering every source through council/council.py. Use for council phase P1, REQUESTs addressed to |
Similar rigs
leavemagic-cyber/earned-confidence
A behavior contract for AI coding agents — every rule added after a real failure. Ships with a verifier agent that cannot edit what it judges.
Pragmatist 3.6k tok ·
frontier-infra/adl
Agent Discipline Layer (ADL) — keep a coding agent honest and make it prove its work: layered guardrails + a /goal contract Warden verifies against reality and signs. Claude Code today; Grok/Codex CLI/Pi planned. (formerly Claude Layers)
Pragmatist 1.2k tok ·
j0yen/dotfiles
The claude-self cluster: machine-personal CLAUDE_SELF.md, its SessionStart hook, and the claude-self CLI.
YOLO Cowboy 4.5k tok ·
witekbobrowski/dotfiles
🗂 Automated custom macOS config
Orchestrator 988 tok ·