Caveat verdict
huo15-img-prompt
An AI image prompt generation skill with 88 aesthetic presets and artist anchors; network_in is for fetching reference images and agent_memory is for storing learned style presets, all consistent with the stated creative purpose.
⚠ Flagged for review — coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.
Automated static analysis — not a human review. Caveat flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
Permission integrity
agent_memory
Findings (8)
HTTP request to bare IP address — common in malicious payloads
SKILL.md · code · http://127.0.0.1
Pipe to bash — executes piped content as shell commands
examples/README.md · prose · downgraded · | Bash
Pipe to python — executes piped content as Python code
examples/recipe-1-brand_kv.sh · prose · downgraded · | python3
Accesses shell history/config
QUICKSTART.md · code · ~/.zshrc
<script> tag in markdown — potential code injection
scripts/storyboard.py · prose · downgraded · <script>
Python urllib.request — network access
scripts/auto_iterate.py · prose · downgraded · urllib.request
Python os.environ.get — reads environment variable
scripts/auto_iterate.py · prose · downgraded · os.environ.get(
Blob URL — may embed executable content
scripts/reverse_prompt.py · prose · downgraded · blob:
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class A/B). Final tier capped at Caution — cannot be lifted by any downgrade, example-payload opt-in, or allowlist.
Permissions & capabilities
No declared permissions — minimal attack surface.
network_inagent_memory Is this flag fair?
Thanks — recorded.