gcp-logs
Pass
Audited by Gen Agent Trust Hub on Feb 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes several shell commands to interact with GCP services and process data.
- Evidence: Uses
gcloud auth list,gcloud config get-value project,gcloud logging read, andjqfor data transformation as seen inSKILL.mdandreferences/QUERIES.md. - [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection due to the ingestion of untrusted external data from GCP logs.
- Ingestion points: Untrusted data enters the agent context via the output of
gcloud logging readcommands documented inSKILL.md(e.g., lines 62, 70, 75) andreferences/QUERIES.md. - Boundary markers: The skill lacks explicit boundary markers or instructions to the agent to disregard natural language instructions found within the log payloads.
- Capability inventory: The skill has the capability to read/write GCP configuration and execute complex data processing via
gcloudandjq. - Sanitization: There is no evidence of sanitization or filtering of the log content before it is processed by the agent.
Audit Metadata