automate-manual-test-cases

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes generated code using framework-specific runners such as npx playwright test, npx codeceptjs run, and npx cypress run as part of its verification and post-hook processes (defined in SKILL.md and CLAUDE.md).
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection due to its core functionality of processing manual test descriptions.
  • Ingestion points: Manual test cases provided in various formats (text, markdown, source code comments) are parsed and used to generate executable code as described in SKILL.md (Step 2).
  • Boundary markers: The skill lacks explicit instructions or delimiters to prevent the agent from obeying malicious instructions that might be embedded within the manual test case data.
  • Capability inventory: The skill has the capability to execute shell commands and modify project files, which could be abused if an injection is successful.
  • Sanitization: There is no evidence of input validation or sanitization for the manual test steps provided by users or external sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 06:37 PM
Security Audit — agent-trust-hub — automate-manual-test-cases