feedback_submit

Management

Submit Scenario Feedback

Send the user's explicit feedback to Scenario via PostHog. Only call after the user chooses to share their reviewed objective and answers; never infer sentiment or outcome, include transcripts, or submit on their behalf without that choice. No credits are spent. For confirmed answers in chat, call this tool directly and omit feedback_id; do not call feedback_request or open a card to obtain an ID. Args: optional feedback_id (omit for a new chat response; use the card ID for app submissions); objective (max 500 characters); sentiment satisfied/mixed/frustrated; outcome yes/partly/no; optional feedback_comment (max 2000 characters); source conversation (default) or app. Context IDs are optional. Returns: status received after ingestion acknowledges the response, or an error. The tool generates feedback_id when omitted and returns it in text and structuredContent on success or an ingestion error. Reuse that ID and the same answers for retries; omit it for a new response. If a transport failure loses the whole result and no ID is known, delivery is unknown: do not automatically resubmit. Example: feedback_submit({objective: "Create game icons", sentiment: "mixed", outcome: "partly", feedback_comment: "The style drifted between icons"}).

open-world

Parameters

Name	Type	Required	Description
team_id	string	—	Team ID. Required if user belongs to multiple teams.
project_id	string	—	Project ID to scope the operation to.
response_format	enum(json \| markdown)	json	Output format: 'json' for structured data, 'markdown' for human-readable text.
feedback_id	string	—	Optional response UUID. Omit for a new chat submission: the tool generates and returns it. Use the card ID for app submissions, or the returned ID when retrying. No feedback_request call or client-generated UUID is needed.
objective	string	✓	Brief objective in the user’s words. The user must review it before submission; omit private details.
sentiment	enum(satisfied \| mixed \| frustrated)	✓	User-selected sentiment; never infer it from their conversation.
outcome	enum(yes \| partly \| no)	✓	Whether the user says their objective was achieved.
feedback_comment	string	—	Optional improvement suggestion explicitly shared by the user. No transcripts or credentials.
source	enum(app \| conversation)	conversation	Where the user submitted the response.