pg-aiguide

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Beyond the annotations (readOnly, idempotent), the description discloses the hybrid search mechanism, BM25, semantic vectors, RRF fusion for mid values, and supported database platforms. This adds substantial behavioral context without contradicting annotations.

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?

The description is two sentences, front-loaded with the core purpose, and every piece of information earns its place. No fluff or redundancy.

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?

Given the presence of an output schema and detailed annotations, the description sufficiently covers the tool's functionality, parameter tuning, and supported sources. It does not need to detail return values since the output schema exists.

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?

The schema already covers all parameters, but the description adds meaning by explaining the semanticWeight scale and that query is used for both BM25 and embedding where relevant. This goes beyond simple parameter names.

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?

The description clearly states the tool searches documentation using hybrid semantic and keyword search, which is a specific verb+resource combination. It also distinguishes from the sibling tool view_skill by focusing on search rather than viewing.

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?

The description provides explicit guidance on using semanticWeight values (0, 1, blend) and lists supported sources. However, it does not explicitly contrast with view_skill or state when not to use this tool, so it lacks full alternative guidance.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

The annotations already declare readOnlyHint=true and idempotentHint=true, which informs the agent that this is a safe, read-only operation. However, the description adds no behavioral detail beyond that—it does not mention how the tool returns skill contents (e.g., file text, directory listings), error cases, or any other side effects. The massive skill list is reference data, not behavior disclosure.

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?

The description is extremely long, dominated by the available_skills list. While the list is useful, it is not concise and pushes the actual tool purpose statement to the periphery. The phrase "Retrieve detailed skills..." is front-loaded, but the list becomes noise for an agent that just needs to know how to invoke the tool. The instruction that every sentence should earn its place is violated by the sheer volume of repetitive skill descriptions.

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?

For a read-only browsing tool, the description is largely complete: the schema covers both parameters, the output schema presumably describes return values, and the available skills list gives comprehensive context for valid skill_name values. The only gap is the misleading "TimescaleDB operations" scope, which is corrected by the list itself, and the lack of guidance for distinguishing this tool from search_docs. Overall, the tool is well-specified despite the verbosity.

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?

The schema provides descriptions for both parameters (100% coverage), but the description adds substantial value by enumerating all valid skill names with their descriptions and trigger conditions. This effectively serves as an inline enum for skill_name, far exceeding the schema's generic "name of the skill to browse". The path parameter is already well-described in the schema, and the description does not detract from that.

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?

The description clearly states the tool's action: "Retrieve detailed skills for TimescaleDB operations and best practices." This is a specific verb+resource, but the qualifier "for TimescaleDB" is slightly misleading since the available skills cover general PostgreSQL, PostGIS, pgvector, and other topics. The tool's name and title further clarify its intent, and the extensive skill list provides concrete examples of what it can retrieve.

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?

The description does not explicitly state when to use this tool versus alternatives like search_docs. There is no guidance on when to prefer viewing a skill versus searching documentation, nor any exclusions or prerequisites. The available skills list contains each skill's "Trigger when user asks to" sections, but those are usage guidelines for the skill content, not for the view_skill tool itself.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.