Record why a recommendation was or was not a fit; this sharpens future recommendations and the persona
| Name | Required | Description | Default |
|---|---|---|---|
| feedback_types | Yes | One or more supported feedback tags | |
| recommendation_id | Yes | Recommendation id | |
| interaction_source | No | Where the feedback came from in the UI or agent flow |
| Name | Required | Description | Default |
|---|---|---|---|
| data | Yes | ||
| meta | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already indicate it is a non-read-only, non-idempotent operation, but not destructive. The description adds a side-effect statement: 'this sharpens future recommendations and the persona', which discloses that it affects future behavior. However, it omits any mention of auth requirements, rate limits, or details about what happens to the recommendation itself (e.g., does it also dismiss it?), which with the low annotation bar leaves moderate value.
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, well-structured sentence with no filler. It front-loads the purpose ('Record why') and adds a consequence clause that explains the value. Every word 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?
For a simple mutation with high schema coverage and an output schema, the description covers the core action. However, it lacks guidance on how this relates to sibling tools (e.g., whether feedback should precede or replace dismiss), and it doesn't clarify the effect on the recommendation's state. This leaves an agent potentially unsure about the intended workflow.
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 parameters have descriptions. The description adds minimal specific parameter meaning—only that feedback_types captures 'why a recommendation was or was not a fit', which is already implied by the enum. The baseline of 3 is appropriate since the schema carries the primary semantic load.
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 verb 'Record' and the resource 'why a recommendation was or was not a fit', indicating a mutation that logs reasoning. It distinguishes from sibling tools like dismiss or block by focusing on capturing reason rather than performing an action on the recommendation, though it doesn't explicitly name the difference.
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?
No explicit guidance is given on when to use this tool versus alternatives such as signals.recommendations.dismiss, block, or watch. The description implies it is for recording reasoning but does not state conditions, prerequisites, or when not to use it, leaving the agent to infer.
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.
The tools are largely distinct due to the domain-prefixed naming (inbox, jobs, signals, etc.) and detailed descriptions. While there is some overlap among inbox actions like acknowledge, apply_suggestion, and decline, the descriptions clarify each behavior. The signals and recommendations sub-groups also have clear boundaries, making misselection unlikely.
All tools follow a consistent pattern of domain.entity.action or domain.action (e.g., inbox.list, jobs.interviews.add, signals.recommendations.dismiss). This uniform camelCase-with-dots convention makes the set highly predictable and easy to navigate.
At 55 tools, the server far exceeds the 25-tool threshold for 'too many' as per the calibration. While the broad domain of job search management justifies numerous operations, the count is still overwhelming and could overwhelm agents or cause selection errors. Several signal-related tools could potentially be consolidated without compromising functionality.
The tool set covers the full lifecycle of job applications: inbox management (list, get, draft, send), job tracking (add, update, archive, delete), interviews (add, update, delete), offers (create, update, accept, decline, negotiation), and company signals (track, pause, recommend, block). There are no obvious missing operations for the core workflows.
