deploy
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- Command Execution: The skill executes build and deployment tasks using
npm run buildandpa app push. These are standard operations for a CI/CD-focused skill. The use ofnpx --no-installfor thepacommand is a safety-oriented pattern that ensures the tool is executed from the local project environment without downloading unauthorized packages. - Indirect Prompt Injection Surface: The skill reads project context from a
memory-bank.mdfile (Step 1). While this is a standard workflow for maintaining project state, it represents a potential surface for indirect prompt injection if the file contains instructions intended to influence agent behavior. - Ingestion points: The agent reads
memory-bank.mdduring the 'Check Memory Bank' step. - Boundary markers: No explicit delimiters or warnings to ignore embedded instructions are present in the skill's logic.
- Capability inventory: The skill possesses
Bashexecution capabilities (for building and pushing apps) andWritecapabilities (for updating the memory bank). - Sanitization: There is no explicit sanitization of the content read from the metadata file before it is used to inform the workflow.
Audit Metadata