threat-model-feedback

Collect feedback about the most recent threat model and file it as a GitHub issue using the Feedback issue template.

Ask the user these questions one at a time:

1. How useful was this threat model? Options: Very useful / Somewhat useful / Not useful

2. Which section was most valuable? Options: Executive summary / Architecture overview / Threats / Attack scenarios / Mitigations / Data classification / Other

3. Was anything missing or wrong? (free text, or "nothing")

4. Were there any false positives — threats that don't actually apply? (free text, or "none")

5. On a scale of 1-5, how would you rate the overall quality? (1 = poor, 5 = excellent)

6. Would you use this tool again? Options: Yes / Maybe / No

7. Any additional comments? (free text, or skip)

Filing the Issue

After collecting answers, try these in order — a filed issue is more useful than a link the user might not click, so only fall back to a link if neither tool is available.

1. GitHub CLI (gh), if installed and authenticated: Check with gh auth status. If authenticated, create the issue directly:

gh issue create \
  --repo RedHatProductSecurity/agentic-threat-modeling \
  --title "[Feedback] <one-line summary from the usefulness + rating answers>" \
  --label feedback \
  --body "$(cat <<'EOF'
**How useful was this threat model?** {usefulness}

**Which section was most valuable?** {most_valuable}

**Was anything missing or wrong?** {missing}

**Were there any false positives?** {false_positives}

**Overall quality rating (1-5):** {rating}

**Would you use this tool again?** {use_again}

**Additional comments:** {comments}
EOF
)"

Share the issue URL that gh issue create returns.

2. A connected GitHub MCP server, if gh isn't installed or isn't authenticated: If an MCP tool for creating GitHub issues is available in this session, use it against the same repository (RedHatProductSecurity/agentic-threat-modeling) with the same title, labels, and body as above.

3. Fallback: a pre-filled link, if neither of the above is available: Construct a pre-filled "new issue" URL against the Feedback issue form, URL-encoding each answer into its field:

https://github.com/RedHatProductSecurity/agentic-threat-modeling/issues/new?template=feedback.yml&title=%5BFeedback%5D&usefulness={usefulness}&most-valuable={most_valuable}&missing={missing}&false-positives={false_positives}&rating={rating}&use-again={use_again}&comments={comments}

Present it as:

Thanks for the feedback! Click here to file it: Submit Feedback

The issue will open pre-filled with your answers on GitHub. Just review and click "Submit new issue."