An MCP server that provides a local, persistent, self-maintaining knowledge base for AI agents
29 tools with consistent verb-noun naming (create_, list_, update_, delete_, get_, set_). All tools have descriptions (avg ~80 chars, within baseline). Input schemas present for all tools with typed parameters and descriptions. However, output schemas are not documented, LLMs cannot predict response structure. Error handling is absent; no recovery guidance or actionable error messages. Parameter descriptions lack constraints (enums, ranges, formats). No tool annotations (readOnlyHint, destructiveHint). Composition is sound (single responsibility per tool), but chaining support is weak, responses don't guarantee IDs needed by downstream tools.
Create a new fact in the knowledge base
Create a new prompt template in the knowledge base
Create a new resource in the knowledge base
Create a new skill in the knowledge base
Create a new tag in the knowledge base
Delete a fact from the knowledge base
Delete a prompt template from the knowledge base
Delete a resource from the knowledge base
Delete a skill from the knowledge base
Output schemas not documented. LLMs cannot predict response structure, field names, or types. Breaks tool chaining and forces agents to guess what data is returned.
No error handling or recovery guidance. Tools lack actionable error messages. Agents cannot distinguish retryable vs fatal failures or self-correct invalid input.
Parameter descriptions lack constraints. No enums for status/sourceType fields, no ranges for limit/page parameters, no format specs for dates. LLMs hallucinate invalid values.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint). Agents cannot infer which tools are safe to retry or have side effects without reading descriptions.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 51 | <=2025-11-25 | v2 |
Delete a tag from the knowledge base
Get a configuration value from the knowledge base
Get contextual information from the knowledge base based on a query
Get a user preference from the knowledge base
List all configuration values in the knowledge base
List execution logs from the knowledge base
List facts in the knowledge base with optional filtering
List all user preferences in the knowledge base
List all prompt templates in the knowledge base
List resources in the knowledge base with optional filtering
List skills in the knowledge base with optional filtering
List all tags in the knowledge base with optional filtering
Log an execution event in the knowledge base
Set a configuration value in the knowledge base
Set a user preference in the knowledge base
Update an existing fact
Update an existing prompt template
Update an existing resource
Update an existing skill
Update an existing tag's name or description
Destructive operations (delete_*) lack confirmation/dry-run support. Agents can permanently delete data without safeguards or reversibility.