Caveat verdict
github-sync-skill
Accesses agent memory/notes AND makes external network calls
Incrementally syncs local Claude skills to a GitHub repository using GITHUB_TOKEN; behavior is transparent (detect new skills, upload files, update README), all operations go to the user's own repo, and the skill explicitly advises against committing the token.
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.
What it does
These are capability combinations: each listed behavior occurs in the skill, but Caveat detects co-occurrence — it does not verify that one flows into another. Read the code to confirm a live chain.
Accesses agent memory/notes AND makes external network calls — may leak personal data
LLM02 · LLM07 · ASI03 · ASI06
Accesses agent memory AND makes external network calls — may leak conversation history
LLM02 · LLM07 · ASI06
Permission integrity
network_out
agent_memory
Findings (2)
Accesses shell history/config
references/token-troubleshooting.md · code · ~/.zshrc
Possible hardcoded credential
scripts/cleanup_remote_repo.sh · prose · downgraded · TOKEN="${GITHUB_TOKEN:-}
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class B/F). 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.
credential_accessagent_memorynetwork_out Thanks — recorded.