A Model Context Protocol server for the Darkmoon security toolbox, providing health diagnostics, generic command execution, and workflow discovery/execution capabilities.
Two tools with minimal descriptions and no visible input schemas in the provided source. get_session lacks actionable context (what is the session ID used for?). tokenize_prompt has a longer description but no schema validation visible. Both tools are READ_ONLY, which is good, but descriptions fall short of the 50-200 char LLM-optimized baseline. No error handling guidance, no parameter validation rules, and no output schema documentation visible. The Dockerfile dominates the source excerpt, obscuring the actual MCP server implementation.
Return the current MCP session ID. This ID is generated automatically when the server starts. It stays the same for the entire lifetime of the server.
Tokenize a launch prompt BEFORE it reaches the model. The Darkmoon privacy boundary normally acts only on tool calls and their output. The initial campaign prompt (TARGET / CREDS / SCOPE / TOKEN / IP / port) is sent to the model first, so without this it would reach the model in clear (issue #40, section 3). This applies the SAME per-session vault the tool calls use, so every placeholder is identical to the ones the gateway rehydrates during the run and restores into the local report — no value is ever hard-coded.
No input schemas visible for either tool. get_session declares empty input {}; tokenize_prompt input schema not shown in source code excerpt. Cannot verify parameter types, constraints, or descriptions.
get_session description ("Return the current MCP session ID...") is 95 chars but lacks actionable context: when should an LLM call this? What does the session ID enable? Why is it needed before tokenize_prompt?
tokenize_prompt description references issue #40 and internal vault mechanics but does not explain the user-facing benefit or when to call it. LLMs cannot infer 'privacy boundary' or 'placeholder rehydration' without explicit guidance.
No output schemas documented for either tool. LLMs cannot plan downstream calls or extract required fields (e.g., does get_session return {session_id: string} or {id: string, created_at: timestamp}?).
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 50 | 2026-07-28+ | v2 |
No error handling guidance. What happens if tokenize_prompt receives an invalid session_id? What if the vault is not initialized? LLMs receive no recovery path.