Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
rafaeelaudibert
left a comment
rafaeelaudibert
left a comment
There was a problem hiding this comment.
Look good, but let's have @MattBro review this once he's back
|
Vale prose linter → found 0 errors, 8 warnings, 0 suggestions in your markdown Full report → Copy the linter results into an LLM to batch-fix issues. Linter being weird? Update the rules!
|
| Line | Severity | Message | Rule |
|---|---|---|---|
| 19:56 | warning | Use 'X' instead of 'x'. | Vale.Terms |
| 22:150 | warning | Capitalize 'Experiments' for PostHog's product. Use 'experiments' for the general industry concept. | PostHogBase.ProductNames |
| 24:43 | warning | Capitalize 'Experiments' for PostHog's product. Use 'experiments' for the general industry concept. | PostHogBase.ProductNames |
| 36:60 | warning | Capitalize 'Experiments' for PostHog's product. Use 'experiments' for the general industry concept. | PostHogBase.ProductNames |
| 74:26 | warning | Capitalize 'Surveys' for PostHog's product. Use 'surveys' for the general industry concept. | PostHogBase.ProductNames |
| 76:81 | warning | Capitalize 'Surveys' for PostHog's product. Use 'surveys' for the general industry concept. | PostHogBase.ProductNames |
| 106:146 | warning | Capitalize 'Experiments' for PostHog's product. Use 'experiments' for the general industry concept. | PostHogBase.ProductNames |
| 108:48 | warning | Use 'X' instead of 'x'. | Vale.Terms |
MattBro
left a comment
MattBro
left a comment
There was a problem hiding this comment.
A couple of suggestions on prompts that may not work as described given the MCP's current tool capabilities.
|
|
||
| **Suggested prompt:** | ||
|
|
||
| > Create a survey asking users: ‘What’s the one thing you’d change about our product?’ Then group the responses by topic and summarize the main themes. |
There was a problem hiding this comment.
The MCP's survey-stats tool returns aggregated statistics, not raw text responses, so "group by topic and summarize themes" isn't something the agent can realistically do today. The prompt also implies the agent will wait for responses to arrive during the same session, which won't happen.
Suggestion: Scope the prompt to creation only, then mention the analysis as a follow-up in the insights paragraph:
Create an open-text survey asking users: 'What's the one thing you'd change about our product?' and target it to users who've been active for at least 30 days.
Then the insights paragraph could add: "Once responses come in, use a follow-up prompt to pull survey stats and identify the most common themes."
4 participants
Wrote the v0 launch blog, planning to publish on Feb 18