Caveat verdict
memory-pipeline
Agent memory pipeline extracting facts from local session logs and generating briefings; LLM API calls use auto-detected user API keys for documented summarization.
⚠ 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.
Findings (5)
Possible hardcoded credential
references/setup.md · code · API_KEY="sk-ant-...
pip3 install — installs Python packages at runtime
references/setup.md · code · pip3 install
References agent memory files
references/setup.md · code · daily-note
Python os.getenv — reads environment variable
scripts/memory-briefing.py · prose · downgraded · os.getenv(
POSTs data to external URL
scripts/memory-briefing.py · prose · downgraded · .post(
"https://
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class 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.
Is this flag fair?
Thanks — recorded.