base44-troubleshooter
Pass
Audited by Gen Agent Trust Hub on Feb 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Executes shell commands to interact with the project environment and the Base44 platform.
- Evidence: Uses npx base44 whoami to verify identity, cat base44/.app.jsonc to read local configuration, and npx base44 logs to retrieve function logs.
- Context: These are standard operations for the base44 troubleshooter skill and originate from the tool's vendor.
- [EXTERNAL_DOWNLOADS]: Utilizes npx to execute the base44 package, which may be downloaded from the npm registry.
- Evidence: Frequent use of npx base44 in SKILL.md and references/project-logs.md.
- Context: This is the intended delivery mechanism for the vendor's CLI utilities.
- [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing external log data.
- Ingestion points: Production logs fetched via npx base44 logs.
- Boundary markers: Absent. Log output is not wrapped in specific delimiters to prevent the agent from interpreting log content as instructions.
- Capability inventory: The agent has access to local file reading (cat) and the vendor's CLI tool (npx base44).
- Sanitization: Absent. The logs are presented as raw text output.
- Context: This risk is inherent to the skill's primary purpose of log analysis and is considered acceptable for this specific use case.
Audit Metadata