writing-manual-test-cases

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from Jira tickets, Confluence requirements, and GitHub Pull Requests, creating a potential surface for indirect prompt injection.
  • Ingestion points: External data is ingested via the bitwarden-atlassian-tools skill and GitHub CLI commands (gh pr view, gh pr diff).
  • Boundary markers: The instructions include a specific defensive directive to treat all ingested content as untrusted data rather than instructions, explicitly citing CWE-1427 and instructing the agent to ignore imperative text.
  • Capability inventory: The skill is authorized to write test case files (Write) to a plugin-specific data directory and execute scoped GitHub CLI commands via Bash.
  • Sanitization: The skill relies on prompt-based isolation to prevent malicious instructions in the source data from influencing agent behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 06:11 PM
Security Audit — agent-trust-hub — writing-manual-test-cases