jfrog-oidc-setup
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill interpolates untrusted user inputs (
github_repos,project_key,github_host) directly into JSON payloads forjf apicalls. - Ingestion points: The
github_repos,project_key, andgithub_hostinputs are used to construct request bodies for the JFrog Platform API in SKILL.md. - Boundary markers: The skill lacks explicit boundary markers or instructions to ignore embedded control characters when processing these strings inside the shell commands.
- Capability inventory: The skill possesses the capability to perform authenticated network operations via the
jf apitool, including creating OIDC providers and identity mappings on the JFrog Platform. - Sanitization: While the skill performs basic sanitization by replacing slashes with hyphens in mapping names (using
tr '/' '-'), it does not escape characters like double quotes or backslashes. This creates a surface for schema confusion where a malicious or malformed input could inject additional fields into the JSON payload sent to the JFrog API.
Audit Metadata