Python Code Validator

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

The description goes far beyond the boolean annotations by disclosing side effects (code execution), sandbox limitations (no network, read-only filesystem, timeout), authentication/payment requirements (HTTP 402, xDAI credits, x402 fallback), and subtle behaviors (fixed_code null semantics, repaired code runs exactly once, examples/expected_output diagnostics). No contradiction with 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 functionally rich but overly long, repeating some schema-level parameter details and including extensive edge-case instructions. It is organized with clear labels (Auth, Arguments, Returns) and front-loads the core purpose, but would benefit from trimming redundant parameter explanations to improve readability.

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?

The description covers all essential contextual aspects: side effects, sandbox restrictions, auth, parameter nuances, error handling (402, 400, 413), output schema pointer, and data retention policy. Given the tool's complexity, the description is complete enough for an agent to select and invoke it correctly.

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?

Even with 100% schema coverage, the description adds significant value: byte-length measured after UTF-8 encoding, 400/413 errors, line numbering starting at 1, deployment-specific timeout cap of 30 vs schema's 60, and clarifications on examples/expected_output behavior that inform parameter usage.

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 function: it performs everything repair_python does and then RUNS the code in a throwaway container, reporting exit code, stdout, and stderr. It also explicitly distinguishes itself from siblings validate_python (diagnosis only) and repair_python (fix only), noting neither runs code.

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 tells when to use it: 'Use it when you need proof that the code runs, or that it does what it says.' It names alternatives and their non-execution nature, and provides parameter guidance such as raising max_iterations for files with multiple faults and setting expected_output or examples to verify correctness.

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 description goes well beyond annotations by detailing authentication requirements (paid key, HTTP 402, x402 payment flow), clarifying that the code is never run, explaining the null fixed_code semantics, and listing which options are silently ignored. This enriches the annotation hints without contradicting them.

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 long but every sentence carries essential information. It is well-structured across purpose, auth, arguments, and returns, with no fluff or repetition. Front-loading the core purpose and key caveat ('code is never run') makes it easy to scan.

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 tool of this complexity, the description covers all critical aspects: behavior, auth, parameter semantics, ignored options, return value summary, and null handling. It leverages the output schema reference where appropriate and leaves no significant gaps for an agent to discover by trial and error.

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?

Even though schema coverage is 100%, the description adds meaningful context beyond the schema, such as advice on when to raise max_iterations, clarifying that transpile_to operates on the repaired source, and explicitly stating that timeout_s, examples, and expected_output have no effect here. This is valuable extra semantic guidance.

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 what the tool does: validation plus deterministic fixes, returning corrected source in fixed_code. It distinguishes itself from siblings by explicitly noting it never runs the code and by framing its output as the fix rather than the diagnosis.

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 provides explicit when-to-use guidance: 'Use it when validation failed and you want the fix rather than the diagnosis.' It also names concrete alternatives (validate_python when diagnosis is enough, execute_python when the fix must be proven to run), which is exactly the recommended level of 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?

Annotations declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false, which the description reinforces without contradiction. It adds substantial behavioral context beyond the annotations: notes the tool is safe on untrusted code, warns that ignored options are not refused, states that unparsable code is answered rather than refused, and discloses data retention ('code and its verdict are retained'). These details materially affect tool use and are not visible from annotations alone.

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 long but front-loaded: the first sentence states the core purpose, followed by usage guidance, auth details, argument semantics, and return pointer. Each section earns its place, though the auth and payment explanation is verbose and could be condensed without losing critical information. Overall, the structure is clear and appropriately detailed for the tool's complexity.

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?

The description covers all critical contextual aspects: what checks are performed, safety profile, usage timing, alternatives, authentication and payment requirements, rate limits, parameter limitations, error behavior, data retention, and a pointer to the output schema. Since an output schema exists, the description need not enumerate return fields in detail, but it references it. This is complete for a complex validation tool.

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 the baseline is 3. The description adds meaning beyond the schema by clarifying that only transpile_to among the options takes effect in static mode, while others (timeout_s, max_iterations, optimize, examples, expected_output) require a pass that rewrites or runs code. It also explains that line/column numbers count from 1 in the submitted fragment, which enriches the code parameter semantics.

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: 'Check Python source without running it: parse, lint (ruff), type-check (mypy), AST security policy, credential scan.' It clearly distinguishes from siblings by listing alternatives: repair_python for corrected source, execute_python to prove code runs. This makes the tool's purpose unmistakable and differentiates it from related tools.

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 when-to-use guidance: 'Use it on every Python file you generated or edited, before writing it to disk.' It also names alternatives with their distinct purposes ('repair_python to get the corrected source instead of the diagnosis; execute_python to prove the code runs') and explains which options are ineffective in this mode, preventing misuse.

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