coleam00/Archon

by coleam00

The first open-source harness builder for AI coding. Make AI coding deterministic and repeatable.

Content Signals

Repo Health

Multi-platform bonus: +5 pts if tool supports 2+ platforms. Score derived from 12 structural signals — not stars or popularity.

Trust & Verification

Risk Assessment

• Multi-agent spawning and orchestration: Repository contains extensive workflow definitions (.archon/workflows/) that spawn multiple AI agents in sequence and parallel, enabling autonomous multi-agent coordination without explicit human approval gates between stages
• Autonomous code execution without human approval: Workflows automate full development cycles (planning → implementation → validation → PR creation) with AI agents making autonomous decisions at each step
• Shell/script execution capabilities: Repository includes .archon/scripts/ with TypeScript and shell scripts (marketplace-security-scan.ts, maintainer-standup scripts) that execute system commands and interact with git/GitHub APIs
• Git worktree management and repository modification: Workflows create isolated git worktrees and perform autonomous git operations (commits, branch creation, PR submission) without explicit per-action human approval
• Test fixtures for malicious code patterns: .archon/scripts/__tests__/fixtures/malicious/ contains intentional examples of credential leaks, reverse shells, RCE, shell injection, and data exfiltration - suggesting the tool is designed to handle or execute potentially dangerous code patterns
• Persistent configuration in repository: .archon/config.yaml and .archon/commands/ directory establish persistent automation configuration that modifies development behavior across all invocations