behavior-mcp
llms.txt24 starsChanged 47 days agoMITView on GitHub
High-Frequency (~60Hz) In-Browser Behavior Tree Execution Engine for AI Agents @putervision/behavior-mcp executes deterministic behavior trees directly in browser runtimes at 60Hz. It bridges cognitive reasoning decisions with deterministic low-latency action sequences, featuring reactive interrupt triggers, frame recordings, telemetry analytics, and multi-layer safety guardrails. Non-claims statement: Functional agency and closed-loop automation — not consciousness, sentience, or AGI.
In this repository15 other files it gives its agents.
- .agents/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- .cursor/rules/agent-reasoning-mcp.mdcCursor rule
- .cursor/rules/behavior-mcp.mdcCursor rule
- .cursor/rules/state-memory-mcp.mdcCursor rule
- .cursor/rules/vision-memory-mcp.mdcCursor rule
- .cursor/rules/world-model-mcp.mdcCursor rule
- agent-reasoning-mcpSkill
- behavior-mcpSkill
- state-memory-mcpSkill
- video-ingestSkill
- vision-memory-mcpSkill
- webcrypt-mcpSkill
- world-model-mcpSkill
The file
Files
# @putervision/behavior-mcp > High-Frequency (~60Hz) In-Browser Behavior Tree Execution Engine for AI Agents `@putervision/behavior-mcp` executes deterministic behavior trees directly in browser runtimes at 60Hz. It bridges cognitive reasoning decisions with deterministic low-latency action sequences, featuring reactive interrupt triggers, frame recordings, telemetry analytics, and multi-layer safety guardrails. ## Core Capabilities - **60Hz Browser Tick Loop**: Injectable standalone executor running sequences, selectors, decorators, and leaf actions. - **Immutable Behavior Trees**: Registered JSON definitions verified with SHA-256 `tree_hash`. - **Reactive Interrupt Triggers**: Priority preemption with cooldown guards preventing interrupt storms. - **Telemetry & Recordings**: Frame-by-frame capture with adaptive playback and duration metrics. - **5-Layer Safety Stack**: Watchdog monitor, rate limiter (max 60 actions/sec), leaf policy gate, kill switch, and origin allowlists. ## 10 Consolidated MCP Tools 1. `load_behavior`: Inject and start behavior tree execution in browser runtime. 2. `set_parameters`: Update execution parameters on the fly. 3. `get_status`: Query active status, current node path, and tick counters. 4. `abort_behavior`: Immediately halt, pause, or resume execution. 5. `register_trigger`: Configure priority interrupts with cooldown guards. 6. `replay_recording`: Capture and replay deterministic frame actions. 7. `get_metrics`: Query execution telemetry and duration statistics. 8. `manage_behaviors`: Register and version behavior tree definitions with SHA-256 tree hashes. 9. `manage_blackboard`: Read and write behavior tree blackboard state variables. 10. `manage_runtime_db`: Database maintenance, diagnostics, and SHA-256 Merkle audit verification. *Non-claims statement: Functional agency and closed-loop automation — not consciousness, sentience, or AGI.*
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.