Caveat verdict
deep-research
deep-research-v7
Chinese-language deep research skill that fetches papers from arXiv/PubMed/PMC and generates structured reports; executionSinkDetected reflects documented bash scripts for paper retrieval and PDF parsing, which are legitimate for the stated research 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.
Findings (7)
Unicode homoglyph detected — uses lookalike characters to evade pattern matching
sources/card-arxiv-test.json · prose
Possible hardcoded credential
scripts/deep_research_v8.py · prose · downgraded · api_key='bce-v3/ALTAKSP-MhjY1gIGq8XyzO87nI28J/60019ccc75714796af69ebaac1ed805a73
subprocess execution — runs system commands from Python
scripts/batch-fetch.py · prose · downgraded · subprocess.run(
Long base64 string (100+ chars) — likely obfuscated payload
sources/card-arxiv-test.json · prose · downgraded · nTheselimitationsmotivateamechanismthatletseachlayerselectivelyaggregateinformat
Python urllib.request — network access
scripts/deep_research_v8.py · prose · downgraded · urllib.request
Python os.environ.get — reads environment variable
scripts/synthesize.sh · prose · downgraded · os.environ.get(
Popular HTTP library — network access
scripts/test-integration.sh · prose · downgraded · got
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class D). 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.