Shared knowledge cache for AI coding agents — reuse an answer once it exists, before your model spends tokens.
Scored across 1 tool
With only one tool, there is no possibility of confusion or misselection between tools. The single tool's purpose—looking up a programming knowledge cache—is clearly stated and unambiguous.
The lone tool 'aieph_search' follows a clear prefix+snake_case convention that is readable and predictable. There are no competing conventions to create inconsistency.
A single read-only search tool is defensible for a narrow lookup-cache service, but it is thin—no get-by-id or topic-listing operations exist. It sits just below a well-rounded tool surface.
Search covers the core read use case for a knowledge cache, but there is no way to fetch a specific cached entry or enumerate categories/topics. These are minor gaps an agent could likely work around via search queries.
Look up the shared programming knowledge cache for error meanings, library usage, and how-tos. Do not send user code, files, or chat history—general questions only. Privacy: queries are PII-masked and logged anonymously. Details: aieph.dev/privacy
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | ||
| session | No | ||
| language | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full behavioral burden and does disclose non-obvious traits: queries are PII-masked, logged anonymously, and there is a content-submission restriction with a privacy-policy link. It stops short of describing return format, caching behavior, or any auth/rate-limit details.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Three short sentences: purpose first, then the submission constraint, then privacy. Nothing is redundant and the most decision-relevant information is front-loaded.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Three parameters with zero schema documentation and no output schema means the definition should explain query/session/language but does not. For a tool an agent must parameterize correctly, the description is incomplete on the invocation side even though its usage and privacy framing are strong.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 0% and the description never mentions the query, session, or language parameters. The ambiguity of 'session' (is it optional context or a cache key?) and 'language' (programming language vs. natural language) is left entirely unresolved, so the description fails to compensate for the coverage gap.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb ('look up') and resource ('shared programming knowledge cache') and enumerates the scope of queries it accepts: error meanings, library usage, and how-tos. An agent can immediately tell this is a read-only, question-answering lookup rather than a code or data operation.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Gives explicit content guidance — general questions only, do not send user code, files, or chat history — which is the key gating condition for correct use. It does not name alternatives or exclusions by tool, but there are no sibling tools, so this is close to complete for the available context.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
aieph_search
