ClawAudit verdict
browser
bsession
Reads local files AND makes external network calls
Browser automation skill orchestrating a Docker-based bsession environment for scripted browser automation; all operations are local and the skill focuses on legitimate session management and script scaffolding.
Automated static analysis — not a human review. ClawAudit flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
What it does
These are capability combinations: each listed behavior occurs in the skill, but ClawAudit detects co-occurrence — it does not verify that one flows into another. Read the code to confirm a live chain.
Reads local files AND makes external network calls — the capabilities for data exfiltration co-occur (data-flow not verified)
LLM02 · LLM06 · ASI03
Accesses credentials AND makes external network calls — potential credential theft
LLM02 · ASI03
Permission integrity
network_out
file_read
credential_access
Findings (5)
Pipe-to-shell pattern (curl | sh) — supply chain attack vector
scripts/install.sh · prose · downgraded · curl -LsSf https://astral.sh/uv/install.sh | sh
Pipe to sh — executes piped content as shell commands
scripts/install.sh · prose · downgraded · | sh
Python urllib.request — network access
SKILL.md · code · urllib.request
References webhook/callback URL
SKILL.md · code · webhook_url
Python os.environ.get — reads environment variable
SKILL.md · code · os.environ.get(
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
Requires 1 system binary. (1 elevated: docker).
network_outfile_readnetwork_incredential_access Thanks — recorded.