Request changes on submitted work (job must be SUBMITTED). Moves job back to ACCEPTED so the human can resubmit. Include a clear reason explaining what needs fixing. The human receives a notification. Use approve_completion instead if the work is satisfactory.
| Name | Required | Description | Default |
|---|---|---|---|
| job_id | Yes | The job ID | |
| reason | Yes | Explain what needs to be revised or fixed | |
| agent_key | Yes | Your agent API key (hp_...) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. It reveals the state change (SUBMITTED -> ACCEPTED), the requirement to include a reason, and the notification to the human. This covers the key behavioral traits, though it does not mention potential errors or response format.
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 four sentences, each earning its place: what it does, the state change, the reason requirement, the notification, and the alternative. There is no fluff or repetition, and it is front-loaded with the core purpose.
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 simple state-transition tool with three params and no output schema, the description covers the essential context: precondition, effect, side effect, and alternative. It does not mention the return value, but that is a minor gap for this kind of 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?
The schema has 100% coverage with descriptions for all three parameters. The description reinforces the 'reason' parameter by saying 'Include a clear reason explaining what needs fixing,' but this adds little beyond the schema. The baseline of 3 is appropriate since the schema does the heavy lifting.
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 specific purpose: 'Request changes on submitted work' and explains the state transition ('Moves job back to ACCEPTED so the human can resubmit'). It distinguishes itself from the sibling tool approve_completion by explicitly mentioning it as an alternative for satisfactory work.
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: the job must be in SUBMITTED state. It also names an alternative tool ('Use approve_completion instead if the work is satisfactory'), giving clear direction on choosing between the two related operations.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Most tools have clearly distinct purposes (e.g., get_human vs get_human_profile differ by access level, get_listing vs get_listings by specificity). There is minor overlap between check_humanity_status and get_human (both return verification info), and the deprecated no-op claim_free_pro_upgrade adds slight clutter, but descriptions otherwise disambiguate well.
All 40 tools follow a consistent snake_case verb_noun pattern (e.g., create_listing, get_job_status, start_stream, submit_verdict). No camelCase or mixed conventions exist; even compound actions like make_listing_offer and leave_review fit the established pattern.
With 40 tools, this is far beyond the 16-25 'heavy' range and into the 'too many' category. While the platform covers a broad domain (hiring, listings, streams, escrow, activation), this many tools could be split into smaller focused servers for maintainability and agent selection clarity.
The surface covers the core lifecycle well: search, hire, pay, communicate, approve, review. However, there is no generic cancel_job tool for regular jobs (only stop_stream for streams) and no update_listing to edit an existing listing, leaving small but navigable gaps in lifecycle management.
