ChatGPT MCP Bridge

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

Annotations already declare readOnlyHint/idempotentHint/destructiveHint=false, so the safety profile is covered; the description usefully adds that contents are not read and which metadata attributes come back. Error behavior for a missing or out-of-root path is left unstated.

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, zero waste, front-loaded with the action and payload before the disambiguation and the read-only confirmation. Nothing could be cut without losing signal.

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?

With no output schema, the description correctly enumerates the returned fields, which is the key omission-risk covered. For a simple two-parameter read-only stat call this is nearly complete; only failure-mode behavior is missing.

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 description coverage is 100% and both parameters are self-documenting, so the schema carries the load. The description's only added signal is the 'approved root' framing, which largely repeats the root parameter's own description; baseline 3 applies.

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+resource (inspect a file or directory) and the exact payload returned (type, size, modification time), plus an explicit scope constraint ('in an approved root'). The 'without reading its contents' clause sharply separates it from read_file, so the agent can route without opening a schema.

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?

Explicitly names the alternative and the condition that selects it ('Use read_file for UTF-8 content'), which is real routing guidance. It doesn't address the other plausible siblings (list_files, search_files) or say what to do when the path doesn't exist, so it stops short of full when/when-not coverage.

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?

Annotations already declare readOnlyHint, destructiveHint=false, idempotentHint, and closed-world scope, so the safety profile needs no restating. The description still adds real value with the 1,500-entry cap and the non-recursive 'immediate' scope, which are not in the annotations; it just doesn't say what happens when the cap is exceeded.

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, front-loaded with the core action and scope, then the sibling routing, then the safety note. Every sentence carries distinct information with no repetition.

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?

With no output schema, the description usefully enumerates the returned fields and the entry cap, which is most of what an agent needs. The only gap is truncation behavior past 1,500 entries, which matters for a listing tool but is a minor omission overall.

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 100%, so both 'root' and 'path' are already documented with defaults and meanings. The description's word 'immediate' reinforces that the listing is non-recursive, but adds no path syntax or root-selection guidance beyond what the schema provides, making the baseline 3 correct.

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 opens with a specific verb and resource (list immediate files and directories) and states the scope limits precisely: non-recursive, an approved root, capped at 1,500 entries. It also names the returned fields (relative path, type, size, modification time), so an agent knows exactly what this produces versus search_files.

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?

It explicitly names the alternative, 'search_files for recursive discovery,' which is the key decision point for choosing between the two. The closing 'Does not change files' further signals this is a safe read for inspection-style tasks rather than mutation workflows.

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?

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false, so safety is covered. The description still adds useful value by clarifying the scope of the return: only approved root names and access modes, explicitly excluding filesystem paths and file contents. It does not mention error or empty-result behavior, keeping it below 5.

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?

One sentence plus one clarifying clause, with the core purpose front-loaded and the negative scope statement following. Every clause earns its place.

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?

There is no output schema, so the description must convey return shape; it does so at a high level ('root names and access modes', not paths or contents). What is missing is any hint of result format or ordering, a minor gap for a simple zero-parameter read.

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 tool takes zero parameters, so the schema is trivially complete and there is nothing for the description to disambiguate. Baseline of 4 applies.

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 and resource ('List the approved root names and access modes') and immediately delimits it from what sibling file tools do by noting that filesystem paths and file contents are not returned. An agent can tell this apart from list_files or file_info without opening any schema.

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 sequencing guidance: 'Call this before other file tools to discover valid root values.' That is a clear when-to-use rule tied to sibling tools, but no when-not-to-use case or alternative is named, so it falls short of a full 5.

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?

Annotations already declare read-only, idempotent, closed-world, non-destructive, so the safety profile is free. The description adds genuinely non-obvious behavior beyond that: a concrete truncation ceiling (512 KB default) with a visible truncation marker, UTF-8 decoding assumptions, and binary rejection. It omits error semantics for missing paths or oversized files, keeping it out of the top band.

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?

Two tight sentences with zero filler: the capability and truncation behavior come first, then the alternative routing and exclusion. Every clause carries information an agent needs before calling.

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?

With no output schema, the description usefully characterizes the return value as UTF-8 text with an explicit truncation marker, which is the key thing an agent needs to interpret results. For a two-parameter read tool with full annotation coverage this is nearly complete; error behavior on unreadable paths is the only notable omission.

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 100%, so both 'root' and 'path' are already documented, and no enum constraints exist. The description reinforces the notion of an 'approved local root' but adds no new syntax, format, or constraint information about the two parameters beyond the schema, so the baseline 3 applies.

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 and resource ('Read text from an approved local root') with concrete scope modifiers: UTF-8 decoding, MAX_READ_BYTES truncation, and rejection of binary files. It separates itself from search_files by telling the agent to locate passages there first, but does not distinguish itself from file_info or list_files, which are the closest competing reads.

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?

'Use search_files first to locate passages' gives explicit routing to a sibling, and 'binary files are not supported' is a clear exclusion. It stops short of full when/when-not coverage: it never says when to prefer file_info (metadata only) over read_file, nor what to do when a file is large or missing.

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?

Annotations already declare this as a safe, idempotent, non-destructive read, and the description adds useful behavior beyond them: results are bounded by max_results and a file scan limit, symlinks and build folders are skipped, and results are relative paths with brief snippets. It does not detail error behavior for missing roots or permission failures, but the operational profile is substantially enriched.

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?

Two tightly written sentences with no filler. The core search behavior and scope are front-loaded, and the alternative routing comes after the primary definition.

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?

No output schema exists, so the description appropriately explains the return shape: relative paths and brief snippets, bounded by limits. It also covers traversal exclusions. Minor gaps remain around failure modes, but for a read-only search tool with full schema coverage and complete annotations, it is adequately complete.

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 description coverage is 100%, so all four parameters are already documented, including query matching semantics and max_results bounds. The description reinforces that matching is case-insensitive and substring-based, but adds little beyond what the schema already provides.

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 gives a specific verb and resource ('Recursively search filenames and UTF-8 contents'), plus the matching model ('case-insensitive substring') and scope ('inside an approved root'). It also explicitly contrasts with read_file, so an agent can distinguish it from siblings without opening the schema.

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?

It clearly routes the agent to read_file for complete text rather than snippets, which is an explicit alternative. It does not state exclusions or when-not to use search, but the retrieval-use case is clear enough to support correct selection.

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