OpenParable
Skip to content
← Back to skills

Feedback

Productivity
feedback

Feedback is an Agent Skill for productivity published in the kbanc85/claudia repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Send feedback, ideas, or bug reports about Claudia. Opens a pre-filled GitHub Discussion with system context. Use when user says "feedback", "suggestion", "report a bug", "feature request", or "I have an idea".

View on GitHub ↗
QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

  1. Open your project in Codex.
  2. Copy and paste the prompt below.
  3. Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/kbanc85/claudia/blob/HEAD/template-v2/.claude/skills/feedback/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/feedback, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

Claudia Feedback

Help users share feedback, ideas, bug reports, or feature requests about Claudia. Opens a GitHub Discussion with auto-populated system info so the user just writes their message.

Trigger Phrases

  • "feedback", "give feedback", "send feedback"
  • "I have a suggestion", "feature request", "idea for Claudia"
  • "report a bug", "something's broken", "I found a bug"
  • "how do I report an issue"

Process

Step 1: Ask What's On Their Mind

Ask the user what kind of feedback they want to share. Keep it casual:

  • "What's on your mind? Bug, idea, or just general feedback?"

If they've already described it, skip this step.

Step 2: Gather System Context

Collect system info silently (don't show raw data to user). Run these:

# Version
cat package.json 2>/dev/null | grep '"version"' | head -1

# OS and Node
echo "$(uname -s) $(uname -m), Node $(node -v 2>/dev/null)"

# Memory daemon health
~/.claudia/daemon/venv/bin/python3 -m claudia_memory --preflight 2>&1 | grep "Result:" || echo "Daemon: not checked"

# Memory count
sqlite3 ~/.claudia/memory/claudia.db "SELECT COUNT(*) FROM memories WHERE invalidated_at IS NULL;" 2>/dev/null || echo "0"

Step 3: Categorize

Map their feedback to a GitHub Discussions category:

Feedback TypeCategory
Bug reportIdeas (no Bugs category by default)
Feature request / ideaIdeas
General feedbackGeneral
QuestionQ&A

Step 4: Build the Discussion URL

Construct a pre-filled GitHub Discussion URL. The URL format:

https://github.com/kbanc85/claudia/discussions/new?category={CATEGORY}&title={TITLE}&body={BODY}

The body should include:

  1. The user's feedback (their words, not paraphrased)
  2. A "System Info" section at the bottom with the collected context

Body template:

{user's feedback text}

---

**System Info** (auto-generated by Claudia)
- Version: {version}
- OS: {os}
- Node: {node_version}
- Memories: {count}
- Daemon: {healthy/not running}

URL-encode the title and body for the query string.

Step 5: Open in Browser

Use the Bash tool to open the URL:

# macOS
open "{url}"

# Linux
xdg-open "{url}" 2>/dev/null || echo "Open this URL: {url}"

# Windows
start "{url}"

Step 6: Confirm

Tell the user:

  • "Opened a new discussion in your browser. Add any details and hit Submit."
  • If the browser didn't open, show the URL directly so they can copy it.

Important

  • Never post feedback without the user's knowledge. This opens a browser for them to review and submit.
  • Don't paraphrase their feedback. Use their exact words in the body.
  • Keep system info minimal. Version, OS, memory count. No sensitive data, no file paths, no personal info.
  • The user submits, not Claudia. The browser opens a pre-filled form. The user reviews and clicks Submit.

GitHub Discussions URL

Repository: kbanc85/claudia

Direct link to create a discussion: https://github.com/kbanc85/claudia/discussions/new