verification-before-completion
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of instructional markdown and reference files. Its primary purpose is to define a 'Definition of Done' and provide checklists for manual or automated verification of developer tasks. No malicious scripts, hardcoded credentials, or exfiltration patterns were identified.
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to execute commands and process their outputs, which creates a surface for indirect prompt injection if the output (e.g., test results, logs, or error messages from an attacker-controlled project) contains instructions.
- Ingestion points: Standard output (stdout) and standard error (stderr) from verification tools (SKILL.md, references/verification-checklist.md).
- Boundary markers: The skill suggests quoting or pasting raw output but does not mandate specific security delimiters.
- Capability inventory: Uses 'Bash', 'Read', 'Grep', and 'Glob' to perform file and system operations (SKILL.md).
- Sanitization: None specified; the skill relies on the agent's internal safety filters when processing tool output.
Audit Metadata