pulumi-context-api
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute several CLI commands using the
pulumitool, specificallypulumi api GetGraphQuerySchema,pulumi api GraphQuery, andpulumi whoami -v. These are used to authenticate, retrieve infrastructure metadata, and perform complex graph queries. - [INDIRECT_PROMPT_INJECTION]: The skill utilizes a dynamic 'primer' (schema documentation) fetched from the Pulumi API to define its operational logic and query syntax in real-time. This creates a dependency on external data to guide agent behavior.
- Ingestion points: Output from the
pulumi api GetGraphQuerySchemacommand referenced inSKILL.md. - Boundary markers: Absent; the agent is directed to process the entire API response as a 'contract' for query composition without specified delimiters.
- Capability inventory: The agent can perform network-based API queries via
pulumi api GraphQueryand access account details viapulumi whoami. - Sanitization: No specific filtering or validation mechanisms are described for the content retrieved from the primer before it influences query construction logic.
Audit Metadata