httpie
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates querying external RESTful APIs and processing their responses. This creates a surface for indirect prompt injection where malicious instructions embedded in a remote API response could influence the agent's behavior.
- Ingestion points: Data from external services is ingested into the agent's context through the output of HTTPie commands in
SKILL.md(e.g.,http GET,http POST). - Boundary markers: The instructions do not provide explicit delimiters or warnings for the agent to ignore instructions contained within the external data.
- Capability inventory: The skill enables network requests and the use of environment variables for authentication via the
httptool. - Sanitization: There is no evidence of sanitization or validation of the content retrieved from remote endpoints before it is processed by the agent.
- [EXTERNAL_DOWNLOADS]: The skill uses the HTTPie client to communicate with external domains such as
api.example.comandexample.com. These are utilized as generic endpoints for testing, debugging, and demonstration purposes. - [CREDENTIALS_UNSAFE]: The documentation includes example commands with dummy passwords (
SecretPassword2026,secret). These serve as placeholders to illustrate how to supply credentials to the CLI tool.
Audit Metadata