docker-compose
Pass
Audited by Gen Agent Trust Hub on Nov 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes repository source code and configuration files to determine required services, which constitutes an ingestion point for untrusted data.
- Ingestion points: The agent is instructed to read connection strings, client constructors, ORM/migration configs (e.g.,
alembic.ini,prisma/schema.prisma), and dependency manifests directly from the repository source. - Boundary markers: No explicit delimiters or instructions are provided to the agent to disregard potential embedded instructions within the ingested data files.
- Capability inventory: The skill utilizes
docker composecommands (includingconfig,up,ps, andlogs),docker inspect, andcurlfor system verification and configuration. - Sanitization: The instructions do not mandate specific sanitization or escaping for values extracted from the repository before they are used in generated configuration files or verification shell commands.
Audit Metadata