ClawAudit verdict
server-watchdog
Receives external input AND uses eval
Monitors user-owned remote servers via SSH to check service health and auto-restart crashed processes; all SSH commands target user-specified hosts using user-provided credentials and the behavior matches legitimate DevOps monitoring.
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.
Receives external input AND uses eval — the remote code-injection pattern (data-flow not verified)
LLM01 · LLM05 · ASI01 · ASI05
Findings (6)
Possible hardcoded credential
SKILL.md · code · password:" { send
References sudo — requests elevated privileges
SKILL.md · code · sudo
References child_process — can spawn system processes
scripts/mongodb-watchdog.js · prose · downgraded · child_process
Uses exec() — may execute shell commands
scripts/mongodb-watchdog.js · prose · downgraded · exec(
Node http/https module — low-level network access
scripts/mongodb-watchdog.js · prose · downgraded · require('https')
Accesses sensitive environment variables
scripts/mongodb-watchdog.js · prose · downgraded · process.env.TELEGRAM_BOT_TOKEN
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.
dynamic_evalnetwork_in Thanks — recorded.