planetscale-webhook-automation-recommendations
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates as a policy and recommendation engine. It does not contain executable code, obfuscation, or network exfiltration patterns. It includes instructions to prevent unauthorized modifications to the environment by requiring human-in-the-loop checkpoints for all production-impacting tasks.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing external data via webhooks to generate automation artifacts. However, it incorporates robust security controls to mitigate risks associated with untrusted input.
- Ingestion points: Automated flows ingest PlanetScale event data such as branch.anomaly and branch.schema_recommendation (SKILL.md).
- Boundary markers: Requirements include mandatory signature verification for all webhook receivers using provided secrets (SKILL.md).
- Capability inventory: The skill allows for the creation of development branches, pull requests, and deploy requests, but restricts production deployment or merged changes (SKILL.md).
- Sanitization: The skill advises against posting sensitive data like raw SQL literals or secrets into communication channels and requires DDL testing on non-production branches (SKILL.md).
Audit Metadata