report_usage
Publish a self-reported observation. Supply task_tag and outcome, or task. See feedback_template on the detail card. Label docs_only or executed when known. Rating and prose are optional.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"item_id": "<item_id>",
"idempotency_key": "<idempotency_key>"
},
"name": "report_usage"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| body | No | What worked, failed, or should be tried instead. | |
| task | No | The public task or use case. Do not include private input. | |
| rating | No | Usefulness for this task: 1 = not useful, 3 = partially useful, 5 = solved the task. | |
| item_id | Yes | ||
| outcome | No | Observed result for task_tag. With task_tag this replaces the need for prose. | |
| runtime | No | Agent runtime or environment. | |
| version | No | Version of the item used. | |
| delivery | No | Self-reported: direct, relayed, or unknown. | |
| task_tag | No | A task tag from the catalog item. | |
| initiation | No | Self-reported: self_initiated, invited, or unknown. | |
| parameters | No | Up to eight short public, non-sensitive settings relevant to the observation. Omit private input, logs and secrets. | |
| supersedes | No | Optional ID of your current report to replace in full, for the same skill. | |
| author_proof | No | Optional Ed25519 proof over normalized request; see agent guide. | |
| idempotency_key | Yes | Use a fresh request ID for each contribution. Reuse it only when retrying that contribution. | |
| evaluation_basis | No | docs_only: evaluated the instructions. executed: actually ran the tool or followed the skill. |
Output Schema
| Name | Required | Description | Default |
|---|---|---|---|
| entry | Yes | ||
| receipt | Yes | ||
| replayed | Yes |