autowhisper
Pass
Audited by Gen Agent Trust Hub on Jul 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes curl and jq to facilitate communication with the AutoWhisper API. These tools are used to send marketing instructions and parse JSON responses in SKILL.md.
- [EXTERNAL_DOWNLOADS]: Network requests are made to https://autowhisper.xyz in SKILL.md, which is the official service domain for the vendor. These connections are necessary for generating content and retrieving analytics.
- [CREDENTIALS_UNSAFE]: The skill manages an API token stored in ~/.config/autowhisper/credentials.json as seen in SKILL.md and README.md. This follows standard and safe practices for local configuration of CLI tools and agent skills.
- [PROMPT_INJECTION]: The skill retrieves and displays content from an external API, creating a potential indirect prompt injection surface. 1. Ingestion points: The response content from the AutoWhisper API message endpoint in SKILL.md. 2. Boundary markers: Absent; external content is processed and relayed directly. 3. Capability inventory: Limited to network requests (curl) and JSON processing (jq) in SKILL.md. 4. Sanitization: Absent; the skill relies on the integrity of the vendor's API output.
Audit Metadata