claude-code-fsm-workflow

Listed

Multi-agent workflow for Claude Code where discipline is enforced by hooks, not personas. 23 subagents with strict role separation, context isolation, and nonce-proof reads — plus deterministic AST audit scripts that replace LLM auditors. Agents can't drift, contradict the spec, or claim work done without proof.

plugin 0 stars 0 forks Updated 3 months ago MIT

Install

Plugins install via a marketplace, in two steps.

Step 1

Step 2

View on GitHub

Bundles

Everything this plugin ships — skills, agents, commands, hooks, and MCP servers it bundles.

Hooks (8)

block-map-writes.sh block-model-override.sh block-worker-reads.sh enforce_orchestrate.py nudge_orchestrate.py post_tool_trace.sh surface-map-on-start.sh validate_map_transition.py

Quality Score: 41/100

Stars 20%

Recency 20%

Manifest 20%

Documentation 15%

Issue Health 10%

License 10%

100

Description 5%

100

Details

Author: itsaldrincr
Repository: itsaldrincr/claude-code-fsm-workflow
Created: 3 months ago
Last Updated: 3 months ago
Language: Python
License: MIT

Listed in marketplaces

claude-code-fsm-workflow