audit-fe-api
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources including Sentry production logs and Firecrawl web search results. This exposes the agent to potential indirect prompt injection if malicious instructions are embedded in the logs or searched content.
- Ingestion points:
SKILL.mdStep 1b (Firecrawl search) and Step 2a/2b (Sentry issue/event search). - Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore embedded commands in the external data.
- Capability inventory: The skill has access to local source code via grep, Sentry search tools, and Supabase SQL execution.
- Sanitization: No sanitization or validation of the external content is implemented before processing.
- [DATA_EXPOSURE]: The skill accesses sensitive local and database information.
- Sensitive File Access:
SKILL.mdStep 0d explicitly instructs the agent to grep.envand.env.localfiles for port information. These files are typically used for storing sensitive environment secrets and credentials. - Database Schema Access:
SKILL.mdStep 4c uses thesupabase:execute_sqltool to read table structure metadata from theinformation_schema. - Context: These actions are used for legitimate discovery within the scope of the skill's purpose but represent a potential data exposure risk if the agent is compromised.
Audit Metadata