inbox-to-hubspot-logger
Fail
Audited by Gen Agent Trust Hub on Aug 4, 2026
Risk Level: CRITICALNO_CODEPROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, credential harvesting, or unauthorized data exfiltration were found in the skill's content.
- [NO_CODE]: The skill is entirely composed of natural language instructions and metadata, without any scripts, binaries, or shell commands.
- [PROMPT_INJECTION]: The skill processes untrusted email thread content, which is a surface for indirect prompt injection.
- Ingestion points: Email thread details (sender, recipients, body) as specified in the workflow section of
SKILL.md. - Boundary markers: None present; the skill does not instruct the agent to use delimiters for untrusted input.
- Capability inventory: Proposes CRM log entries and task creation; no direct system-level or network-level execution capabilities exist.
- Sanitization: No explicit sanitization or filtering is described; however, the instruction to 'propose' entries for user confirmation significantly mitigates the risk of unintended actions.
Recommendations
- Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata