MCP Server for Forward Chaining Rete Inference Engine Skill - Pure Python Standard Library implementation of forward chaining rule-based reasoning using Forgy's Rete pattern matching
Two tools with minimal definitions. assert_fact has a basic 3-parameter schema with short descriptions (subject, predicate, object all under 50 chars). query_memory has no parameters and a generic description. Neither tool description explains WHEN to use it, what it returns, or error handling. No output schemas documented. No error guidance. Descriptions are 10-20x shorter than production baselines (194 chars avg). Parameter descriptions lack format constraints, ranges, or examples of valid values. No mention of side effects, idempotency, or prerequisites.
Assert a fact into the working memory of the inference engine
Query the current contents of the working memory
Missing output schemas for both tools. LLMs cannot plan downstream calls or extract data without knowing response structure.
Tool descriptions are 10-50 characters; production baseline is 194 chars. Descriptions lack WHAT, WHEN, and WHAT-IT-RETURNS context.
Parameter descriptions for assert_fact lack format constraints, valid ranges, or examples. 'object' accepts 'any' type with no guidance on what values are valid.
No error handling guidance. If assert_fact fails or query_memory returns empty, LLM has no recovery path.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 36 | 2026-07-28+ | v2 |
assert_fact modifies state but description does not declare this. LLM cannot determine if the call is safe to retry or has irreversible consequences.