Write the final paragraph evaluation file using LLM-supplied editorial strengths, concerns, canon notes, next steps, and verdict explanation combined with objective metrics recomputed from the current chapter state. Call prepare_paragraph_evaluation first to obtain the objective data before calli...
AI agents use write_paragraph_evaluation to create or update resources in Narrarium — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Narrarium environment.
This tool creates or modifies an evaluation file by combining LLM-supplied editorial data with computed metrics. While it produces written artifacts that persist, the operation is not irreversible—evaluation files can be updated, corrected, or deleted without permanent loss. It does not delete or destructively overwrite data, execute arbitrary code, or perform financial transactions.
From the tool's definition Tool description explicitly states 'Write the final paragraph evaluation file' and involves creating/modifying an evaluation document with editorial analysis and metrics. The action is reversible (files can be updated or deleted later).
Attacks that exploit this kind of access
Write the final paragraph evaluation file using LLM-supplied editorial strengths, concerns, canon notes, next steps, and verdict explanation combined with objective metrics recomputed from the current chapter state. Call prepare_paragraph_evaluation first to obtain the objective data before calling this tool. It is categorised as a Write tool in the Narrarium MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Narrarium MCP server in PolicyLayer and add a rule for write_paragraph_evaluation: 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 Narrarium. Nothing to install.
write_paragraph_evaluation 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 write_paragraph_evaluation 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 write_paragraph_evaluation. 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.
write_paragraph_evaluation is provided by the Narrarium MCP server (narrarium-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →