judge_session
Score a dogfood session on 6 dimensions (1-5 each): truth, compression, anticipation, output, delegation, trust. Returns overall score and records failure classes.
This record as markdown: /tools/io-github-homenshum-nodebench/judge-session.md
What judge_session does on Nodebench
AI agents use judge_session to create or update resources in Nodebench, usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Nodebench environment.
Why judge_session is rated Medium
The tool scores/evaluates a session and records results (failure classes, overall score), meaning it writes structured evaluation data to persistent storage. It does not delete data or execute arbitrary code, but it does create/modify records, placing it in the Write category. Misuse could corrupt quality-gate data, hence medium severity.
From the tool's definition Score a dogfood session on 6 dimensions... Returns overall score and records failure classes
Attacks that exploit this kind of access
The rule that runs judge_session safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Nodebench, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For judge_session, this is the rule to start with:
judge_session stays usable, but capped: an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Nodebench, apply this rule, and every judge_session call is checked against it from then on.
Questions about judge_session
Score a dogfood session on 6 dimensions (1-5 each): truth, compression, anticipation, output, delegation, trust. Returns overall score and records failure classes. It is categorised as a Write tool in the Nodebench MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Nodebench MCP server in PolicyLayer and add a rule for judge_session: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Nodebench. Nothing to install.
judge_session is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the judge_session rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for judge_session. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
judge_session is provided by the Nodebench MCP server (nodebench-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Nodebench, and thousands of servers like it.
This server
Across the catalogue