
ACR — Agent Composition Records
An interaction profiling system for AI agents that logs every external tool call and compiles behavioral analytics through queryable lenses. Agents auto-register on first use, then call log_interaction after each MCP server call, API request, or tool invocation to build a friction profile over time. The friction lens shows where your agent loses time and tokens: bottleneck detection, retry waste analysis, chain overhead, and population baselines across all registered agents. You also get anomaly signal notifications when the network observes unusual patterns affecting skills in your agent's composition. Think of it as observability infrastructure for agentic workflows. The SDK works standalone or through MCP tools like get_friction_report and get_network_status. No content logging, just timing and status metadata.
- tools
- 6
- transport
- STDIO
- env var
- 1
- last updated
- Jul 23, 2026
Tools
6 toolsPublic tool metadata for what this MCP can expose to an agent.
orient_meWhere am I, what should I do next — state-aware routing
Where am I, what should I do next — state-aware routing
No parameter schema in public metadata yet.
log_interactionLog an interaction with rich fields (retry_count, chain_id, tokens_used…)
Log an interaction with rich fields (retry_count, chain_id, tokens_used…)
No parameter schema in public metadata yet.
get_friction_reportThe friction lens: where time and tokens go
The friction lens: where time and tokens go
No parameter schema in public metadata yet.
summarize_my_agentEnd-of-session summary
End-of-session summary
No parameter schema in public metadata yet.
get_notificationsUnread anomaly-signal notifications for your composition
Unread anomaly-signal notifications for your composition
No parameter schema in public metadata yet.
get_my_agentIdentity, dashboard link, registration state
Identity, dashboard link, registration state
No parameter schema in public metadata yet.
Sponsors
Configuration
ACR_API_URLACR API endpoint. Defaults to https://acr.nfkey.ai
Readme
ACR — Agent Composition Records
A behavioral registry for AI agents. ACR captures every external tool call your agent makes, compiles those signals into an interaction profile, and shows you where the time went — starting with a readout card at the end of every session.
Get started (60 seconds, Claude Code)
npm i -g @tethral/acr-hook && acr-hook init
That's the whole setup. The hook mints an identity for your agent (no account, no API key to manage), wires itself into Claude Code's tool hooks, and captures every tool call automatically. When your next session ends, a card prints in your terminal:
── ACR session card ──
142 tool calls | 4.1% of active time waiting
Top sinks: platform:bash 38s · mcp:github 12s
Full report: https://dashboard.acr.nfkey.ai/agents/…
Details
- Repository
- github.com/tethral-inc/agentregistry
- Website
- acr.nfkey.ai
- Category
- Monitoring & Observability
- Package
@tethral/acr-mcp- Transport
- STDIO
- Registry
- MCP registry, active