MCP server providing access to corporate security baseline policy documents for cloud architecture auditing. Exposes policy resources and tools for fetching compliance baselines including zero-trust IAM, HA compute rules, and FinOps baselines.
Single tool with basic definition. fetch_policy_baseline has a clear verb-noun name and functional description (89 chars), but lacks critical schema documentation. Input parameter 'policy_name' has a description but no explicit type constraint in the visible schema. No output schema documented. No error handling guidance for the LLM (FileNotFoundError is raised but not mapped to actionable recovery steps). Tool is read-only and safe, but definition quality falls short of production standards due to missing schema formalization and error recovery patterns.
Fetches the corporate security baseline policy document content by name (e.g. 'zero_trust_iam.md' or 'ha_compute_rules.md').
No documented output schema. LLM cannot infer what fields or structure the tool returns, forcing it to guess or make assumptions about downstream usage.
Error handling lacks recovery guidance. FileNotFoundError is raised but not mapped to actionable LLM instructions (e.g., 'Try listing available policies first' or 'Valid names: zero_trust_iam.md, ha_compute_rules.md, finops_baselines.md').
Input parameter lacks explicit type and format constraints in schema. 'policy_name' description mentions examples but no enum or pattern constraint prevents hallucinated filenames.
STDIO transport only. Server is not remotely accessible and cannot be used by hosted MCP clients. Limits deployment to local subprocess spawning only.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 50 | 2026-07-28+ | v2 |
Tool description includes example filenames ('zero_trust_iam.md', 'ha_compute_rules.md', 'finops_baselines.md'). LLMs tend to reuse examples literally; should use enum constraint instead.