Loading ToolBench
Model Context Protocol server for Cymulate — BAS, Exposure Validation, Exposure Analytics, ASM. 100% API coverage (337 endpoints) via 106 tools.
36 tools with consistent naming (verb_noun pattern) and descriptions present. However, most parameter descriptions are generic ('Assessment payload', 'Additional options') without specifying required fields, constraints, or formats. Input schemas use 'object' type with minimal field documentation. Output schemas are not documented. Error handling is absent from descriptions. Tool descriptions average ~80 chars, below the 194-char baseline for A+ tools. No pagination guidance despite tools returning lists. Missing idempotency and retry guidance for WRITE/DESTRUCTIVE operations.
Create a new Advanced Scenarios (Purple Team) assessment.
Get results for a specific Advanced Scenarios assessment.
Get Advanced Scenarios / Purple Team dashboard data.
Get findings from a specific Advanced Scenarios assessment.
Get details of a specific Advanced Scenarios template.
Launch a previously-created Advanced Scenarios assessment.
Generic parameter descriptions ('Assessment payload', 'Additional options') lack field documentation, constraints, or format guidance. LLMs cannot infer required fields or valid values.
No output schemas documented. LLMs cannot plan downstream tool calls or extract required fields (e.g., assessment_id for follow-up queries).
List tools (list_templates, list_assessments, list_history_ids) lack pagination parameters (limit, offset, cursor) and result count guidance. Large result sets risk context window exhaustion.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 52 | <=2025-11-25 | v2 |
List all Advanced Scenarios assessments.
List Purple Team / Advanced Scenarios templates available.
Move a Cymulate agent to a target environment. Uses PUT /v1/agents/change-environment with query params id (agent ID) and envId (target environment ID). Requires the target environment to already exist; use cymulate_env_list to discover env IDs. Requires permission on both source and target envs.
Get all findings for a specific ASM assessment.
Get list of ASM assessment IDs from history.
Get all findings for the latest ASM assessment.
Get detailed technical output from the latest ASM assessment.
Trigger a new Attack Surface Management assessment (external recon).
Get executive report for a specific assessment by ID.
Get the latest executive report results for the selected BAS module.
Get the latest technical report for the selected BAS module.
Get technical (detailed) report for a specific assessment. Use for SIEM/detection engineering.
Get details for a specific template of the selected BAS module.
Launch a Breach & Attack Simulation assessment on the selected module. Requires template_id for most modules; module-specific fields may go in `options`.
List historical assessment IDs for the selected BAS module. Prefer this over the module-specific /ids endpoint.
List templates available for the selected BAS module.
Get current status of the BAS module (running/idle/last-run info).
Stop a running BAS assessment on the selected module.
Create a new launched Exposure Validation assessment.
Delete a specific launched EV assessment.
Get a specific launched Exposure Validation assessment.
Get the launched-assessment configurations available.
Get filters available for launched EV assessments.
List all launched Exposure Validation assessments.
Schedule a new Exposure Validation assessment.
Delete a scheduled EV assessment.
Get a specific scheduled EV assessment.
Get scheduled-assessment configurations available.
List all scheduled Exposure Validation assessments.
Update a scheduled EV assessment.
WRITE and DESTRUCTIVE tools lack idempotency guidance, retry safety, or confirmation patterns. Agents cannot safely retry without risking duplicate assessments or unintended deletions.
No error handling guidance in tool descriptions. LLMs cannot determine if failures are retryable, user-fixable, or fatal. No recovery suggestions provided.