iblai-api-studio-team
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local shell commands to interact with the Open edX Studio API.
- Runs a preflight check via
node .claude/skills/iblai-api-studio-auth/scripts/studio-login.mjsto verify session status. - Uses
curlto perform GET, POST, and DELETE operations against the course team endpoints. - Loads session credentials from a local
.envfile (studio.env). - [DYNAMIC_EXECUTION]: The skill uses a Python one-liner to parse HTML content into JSON.
- Execution pattern:
curl ... | python3 -c 'import sys,re,json; ...'. - The Python script uses a regular expression to extract a JSON array from the
ManageCourseUsersFactoryJavaScript call in the HTML response. - This is a common and legitimate pattern for scraping data from web pages that do not provide a direct JSON API.
- [INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection by ingesting and processing HTML content from the remote Studio server.
- Ingestion points: Reads HTML data from
$STUDIO_URL/course_team/$COURSE(SKILL.md). - Boundary markers: None explicitly defined in the parsing script, though the regex target is highly specific.
- Capability inventory: Performs network operations (
curl) and file writes (via redirection in examples). - Sanitization: Uses standard
json.loadsandre.searchto isolate the specific data array, which limits the influence of arbitrary HTML content on the final output.
Audit Metadata