Skip to content
Back to skills

Feedback

ASecurity

Send feedback to the Nexu team. Use when the user says /feedback followed by their message.

  • 3,264 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added May 26, 2026
toolsshellbashnodeexpressapi

Works with

  • api

Security analysis

A100/100

Pro scans all 2 files and shows the line behind each finding

Scanned May 27, 2026

npx -y skills add nexu-io/nexu --skill feedback --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Feedback?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Feedback
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/nexu-io-feedback/badge)](https://www.skillsdirectory.com/skills/nexu-io-feedback)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: feedback
description: Send feedback to the Nexu team. Use when the user says /feedback followed by their message.
---

# Feedback

Collect user feedback and forward it to the Nexu team. Conversation history and images are extracted automatically by a script — you do NOT need to copy-paste messages or scan for image URLs.

## When triggered

The user sends `/feedback <message>` to share feedback, report issues, or make suggestions about the Nexu platform.

## Steps

1. **Extract feedback content**: The text after `/feedback` is the user's feedback. If empty, ask the user to provide their feedback.

2. **Gather identifiers**:
   - **agentId**: Your own agent ID. Find it from the `Runtime:` line in your system prompt — it appears as `agent=XXXXXX`. It is a cuid2 string like `y9cnvdlucvyaokp20mqrsoa9`. Do NOT use "main" or other placeholder values.
   - **channel**: The current channel type — one of `feishu`, `slack`, or `discord`.
   - **sender**: The sender's display name or username as shown in the conversation. If you only have a user ID, use that. Do NOT use generic "user" or "User".

3. **Run the submit script**: Use the exec tool to run the following command. The script automatically reads your conversation history and any images from the session file — you do NOT need to provide them manually.

```bash
SKILL_PATH="<SKILL_LOCATION>"
node "$(dirname "${SKILL_PATH/#\~/$HOME}")/submit-feedback.mjs" \
  --content "<ESCAPED_FEEDBACK>" \
  --sender "<SENDER>" \
  --channel "<CHANNEL_TYPE>" \
  --agent-id "<AGENT_ID>"
```

Important:
- Replace `<SKILL_LOCATION>` with the exact path from the `<location>` tag in your system prompt (may contain `~`)
- The `${SKILL_PATH/#\~/$HOME}` expansion handles tilde (`~`) → absolute path conversion automatically
- Replace ALL other `<...>` placeholders with actual values BEFORE running the command
- Properly escape shell special characters in the feedback content (single quotes → `'\''`, etc.)

4. **Confirm to user**:
   - If the output contains `{"ok":true}`, reply: "Thanks for your feedback! It has been forwarded to the Nexu team."
   - If it fails, reply: "Sorry, there was an issue sending your feedback. Please try again later."

## Important

- Do NOT modify, filter, or censor the user's feedback content. Forward it as-is.
- Do NOT ask for confirmation before sending — the user already expressed intent by using /feedback.
- Do NOT manually build conversationContext or imageUrls — the script handles this automatically.
- The API will automatically look up the bot owner's email and bot name from the agentId, so focus on getting the agentId right.

Files in this skill

  • SKILL.md2.6 KB
  • submit-feedback.mjs15.8 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.