Documentation
CLI Reference
spark feedback

spark feedback

Rate recommendations from a query session.

Synopsis

spark feedback <session-id> --helpful
spark feedback <session-id> --not-helpful

Description

spark feedback lets you rate the recommendations you received from a spark query session. Feedback is the signal that drives quality in the Spark knowledge network — helpful ratings boost a solution's visibility, while not-helpful ratings reduce it.

Every piece of feedback improves the network for everyone. When you mark a recommendation as helpful, similar queries from other developers are more likely to surface that solution. When you mark one as not helpful, the ranking algorithm adjusts so that better alternatives rise to the top.

AI agents integrated with Spark typically send feedback automatically after applying a recommendation. You can also send feedback manually when reviewing results in your terminal.

Arguments

ArgumentRequiredDescription
session-idYesThe session ID from a previous spark query.

Flags

FlagDescription
--helpfulRate the recommendations as helpful.
--not-helpfulRate the recommendations as not helpful.

Use exactly one of these flags per invocation.

Examples

Mark as helpful

spark query "fix memory leak in Node.js streams" --pretty
# Returns session_id: ses_xyz789
 
# The recommendation worked — send positive feedback
spark feedback ses_xyz789 --helpful

Mark as not helpful

spark feedback ses_xyz789 --not-helpful

Feedback in an agent workflow

When an AI agent uses Spark, the typical flow is:

# 1. Agent queries for a solution
spark query "handle file upload validation in Express"
 
# 2. Agent retrieves the top recommendation
spark insights ses_abc123 0
 
# 3. Agent applies the solution and tests it
 
# 4. Agent sends feedback based on the outcome
spark feedback ses_abc123 --helpful

Related commands