Caveat verdict
viking-girlfriend
Self-hosted AI companion skill using local Ollama and LiteLLM with API keys stored in a local .env file; all model routing is to localhost endpoints with no unexpected exfiltration.
⚠ 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
package_install
Findings (6)
Pipe to bash — executes piped content as shell commands
scripts/model_router_client.py · prose · downgraded · |bash
Pipe to sh — executes piped content as shell commands
scripts/model_router_client.py · prose · downgraded · |sh
Instructs covert action — may act without user awareness
data/core_identity.md · prose · downgraded · quietly
Python os.environ.get — reads environment variable
scripts/crash_reporting.py · prose · downgraded · os.environ.get(
Python shutil file operation — copies/moves/deletes files
scripts/crash_reporting.py · prose · downgraded · shutil.move(
Popular HTTP library — network access
scripts/model_router_client.py · prose · downgraded · got
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class A). 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_inpackage_install Is this flag fair?
Thanks — recorded.