MCP server for MISP threat intelligence platform - IOC lookups, event management, correlation discovery, and intelligence enrichment
Server has 7 tools with reasonable naming (verb_noun pattern) and descriptions present for all. However, parameter descriptions are sparse or missing for many tools. Input schemas are visible but lack depth, many parameters have no description text, violating the pattern:tool-description requirement. Error handling is basic (try/catch with generic messages). No tool annotations (readOnlyHint/destructiveHint) despite clear risk levels. Output schemas are not formally documented. Destructive tools (misp_delete_event, misp_delete_attribute) require confirmation but lack structured guidance for recovery.
Add an IOC/attribute to a MISP event
Add multiple attributes (IOCs) to a MISP event at once
Delete (soft or hard) an attribute from MISP. Destructive: requires confirm:true (or MISP_ALLOW_DESTRUCTIVE=true); hard delete also requires confirmHard:true.
Delete a MISP event (requires appropriate permissions). Destructive: requires confirm:true (or MISP_ALLOW_DESTRUCTIVE=true).
List MISP sharing groups for controlled event distribution
Search for specific attributes (IOCs) across all MISP events
Parameter descriptions missing or incomplete. Many parameters (e.g., 'tags' in misp_search_attributes, 'attributes' in misp_add_attributes_bulk) lack descriptions explaining format, constraints, or expected values. LLMs cannot infer parameter meaning from names alone.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint) despite clear risk levels declared (READ_ONLY, WRITE, DESTRUCTIVE). Spec 2026-07-28 recommends tool annotations for agent safety.
Output schemas not formally documented. Tools return JSON text blobs without declared response structure. LLMs cannot plan downstream calls or extract fields reliably.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 58 | 2025-06-18+ | v2 |
Get MISP server version, status, and diagnostic information
Error messages are generic ('Error getting server status: ...') and do not guide recovery. No actionable next steps or error classification (retryable vs. user-fixable vs. fatal).
Destructive operations (misp_delete_event, misp_delete_attribute) rely on a 'confirm' boolean parameter. No dry-run option or structured confirmation flow. Agents may bypass confirmation if MISP_ALLOW_DESTRUCTIVE=true is set, risking accidental data loss.