hunt-api-misconfig
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides logic for fetching and processing external API specifications and responses, which creates a potential surface for indirect prompt injection from untrusted sources.
- Ingestion points: External API documentation (e.g.,
openapi.json,swagger.json) and HTTP response bodies from target systems. - Boundary markers: Includes mandatory confirmation gates requiring the user to confirm targets, scope, and authorization before any active probing. It also instructs the agent to remain read-only without such confirmation.
- Capability inventory: Uses
curlfor network requests to target APIs andjqfor processing JSON data. - Sanitization: Explicitly recommends schema validation and local processing of specifications to mitigate risks from untrusted data.
- [COMMAND_EXECUTION]: Provides standard diagnostic shell commands such as
curl,jq, andcatfor testing API endpoints and analyzing local files within a security assessment workflow. - [EXTERNAL_DOWNLOADS]: References the
elementalsouls/Claude-BugHunterGitHub repository as a source for technical methodology and documentation. No remote scripts or binary assets are automatically downloaded or executed by the skill.
Audit Metadata