deepseek_run

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

With no annotations, the description carries the full behavioral burden. It discloses positive behaviors: it creates a session, selects DeepSeek, executes, streams progress, and returns a full result plus file audit. However, it remains silent on failure modes, timeout handling, error reporting, whether the call blocks until completion, or what side effects occur in the workdir. For a code-execution tool, these are material gaps, so this is only partially transparent.

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 a single front-loaded sentence that states the action, enumerates the internal steps in sequence, and ends with the deliverable. Every phrase adds meaning; there is no filler, repetition, or vague abstraction. It is concise without sacrificing scope.

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?

This is a complex tool with 10 parameters, no annotations, and no output schema, yet the description is only one sentence. It covers the high-level flow but omits crucial context: preconditions (Haha Desktop, DeepSeek availability), error/failure semantics, synchronous vs asynchronous behavior, what the 'file audit' contains, and how the optional parameters influence execution. For an orchestration tool of this complexity, the description is under-specified.

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 50% (descriptions exist for objective, workdir, modelId, providerId, allowedFiles, but not for notes, constraints, contextFiles, timeoutSeconds, acceptanceCommands). The description itself adds zero parameter-level detail beyond mentioning 'file audit' (tangentially related to allowedFiles). It does not compensate for the undocumented parameters or clarify how timeouts, constraints, or acceptance commands behave. An agent would have to guess at the meaning of half the inputs.

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 uses a precise verb ('Run') and a well-scoped resource ('one complete A-to-B delegation'), then enumerates the concrete steps performed (create session, select DeepSeek, execute, stream progress, return result and file audit). This makes the purpose unmistakable and clearly distinguishes it from the sibling tools (start/wait/cancel/status/profiles), which only cover stages of the pipeline.

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 implies this is the full-pipeline entry point ('one complete A-to-B delegation'), but it never explicitly says when to use this tool versus deepseek_start, deepseek_wait, or deepseek_cancel. It also omits prerequisites (e.g., Haha Desktop running, an active DeepSeek provider) and any exclusions or alternative recommendations. Context is enough for a basic decision, but not explicit.

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