qbr-plan
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted campaign performance data and qualitative client feedback.
- Ingestion points: Campaign performance metrics, goals, actual results, and client satisfaction signals provided by the user as described in SKILL.md.
- Boundary markers: The process instructions do not define specific delimiters or instructions for the agent to ignore potentially malicious content within the provided data.
- Capability inventory: The skill is limited to text analysis and presentation generation; it does not utilize executable code, shell commands, or network operations.
- Sanitization: No methods for validating or sanitizing the input data are mentioned in the skill definition.
- [NO_CODE]: The skill consists entirely of markdown instructions and does not bundle or execute any scripts or binary files.
- [SAFE]: Data access is limited to a specific local marketing directory (~/.claude-marketing/) for profile context, with no evidence of exfiltration or unauthorized system modifications.
Audit Metadata