slack
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWNO_CODEPROMPT_INJECTION
Full Analysis
- [SAFE] (SAFE): No malicious behavior, obfuscation, or data exfiltration attempts were identified. The skill is entirely instructional.\n- [NO_CODE] (SAFE): This skill contains no scripts, binaries, or configuration files, which eliminates risk categories related to remote code execution or malware persistence.\n- [PROMPT_INJECTION] (INFO): The skill is designed to process untrusted user input (messy work updates). While this creates an indirect prompt injection surface, the skill lacks the capabilities (network access, file modification, or command execution) required to exploit such an injection.\n
- Ingestion points: User-provided work notes processed by instructions in SKILL.md.\n
- Boundary markers: Absent in the prompt logic.\n
- Capability inventory: None (Instruction-only; the agent merely formats and displays text).\n
- Sanitization: Absent; the skill relies on the underlying LLM's inherent processing of the input text.
Audit Metadata