rest-graphql-debug
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
- [REMOTE_CODE_EXECUTION]: An automated scan flagged a command where the output of a
curlrequest is piped intopython3. A technical review confirms this is a benign use of thepython3 -m json.toolmodule for formatting API responses and does not represent a code execution threat.- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of data from external API endpoints, which presents a surface for potential indirect prompt injection attacks. - Ingestion points: Examples in
SKILL.mddemonstrate fetching content from remote URLs usingcurland therequestslibrary. - Boundary markers: The skill instructions do not specify the use of delimiters to encapsulate received API data.
- Capability inventory: The skill uses
terminalandexecute_codefor diagnostic tasks. - Sanitization: The skill provides a helper for redacting authentication headers but does not implement content-based sanitization for response bodies.- [CREDENTIALS_UNSAFE]: The skill documentation encourages security best practices by advising against hardcoding credentials and recommending the use of environment variables for token management.
Audit Metadata