Record professional issues, user corrections, and key evidence that routine hooks miss, including observations or counterexamples to improve agent learning.
补充专业问题、用户纠正或关键依据；常规工具事件由Hook采集。
| Name | Required | Description | Default |
|---|---|---|---|
| summary | Yes | ||
| category | No | ||
| event_refs | No | ||
| resolution | No | ||
| knowledge_id | No | ||
| feedback_kind | No | ||
| context_handle | No | ||
| conversation_id | No | ||
| idempotency_key | No | ||
| expected_task_id | No | ||
| knowledge_revision | No | ||
| expected_task_revision | No |
Changes observed during successful MCP inspections.
v0.1.0Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations exist, so the description carries the full behavioral burden. It never says this is a mutation, whether submissions are idempotent, how revision-mismatch conflicts behave, or what happens to previously submitted feedback — significant gaps given the presence of idempotency_key and expected_task_revision in the schema.
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?
A single compact sentence with the primary purpose front-loaded before the exclusion clause — no waste, but the brevity comes at the cost of under-specification rather than being a virtue for a 12-parameter tool.
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 12 parameters, no annotations, and no output schema, the description should explain the submission contract (required summary, optional categorization, conflict/revision semantics). Instead it offers one line, leaving most of the calling contract undocumented.
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 0% across 12 parameters, and the description explains none of them. Ambiguous enums like category (mixed/operation/professional/unknown) and feedback_kind (observation/counterexample) are left entirely undefined, as are the relationship between event_refs, knowledge_id, and context_handle.
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 states the content to submit (professional problems, user corrections, key evidence), which hints at a feedback/case-annotation write tool, but never explicitly says what entity is being supplemented or the effect of the write. It gives no differentiation from siblings like candidate_reference_feedback or knowledge_use, so an agent must infer the distinction.
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 clause '常规工具事件由Hook采集' does provide an implied exclusion: routine tool events are auto-collected, so don't submit them here. However, it never states positively when to call this tool, nor names an alternative sibling for routing.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.