Skills
Navigate
skills/indykite/skills/indykite-ciq-read/Gen Agent Trust Hub

indykite-ciq-read

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill manages a vulnerability surface where untrusted data is ingested for query execution.
  • Ingestion points: scripts/execute.sh accepts a JSON file containing input_params to be passed to the API.
  • Boundary markers: The platform utilizes a $variable syntax within policies and queries to isolate parameters from graph query logic.
  • Capability inventory: The scripts/execute.sh utility performs network requests to the IndyKite REST API via curl.
  • Sanitization: Examples in SKILL.md demonstrate using jq for safe JSON construction, reducing the risk of malformed payload injection.
  • [COMMAND_EXECUTION]: Provides a helper script scripts/execute.sh that uses curl to interact with the IndyKite API. This is expected functionality for the skill's stated purpose.
  • [DATA_EXFILTRATION]: Enables network communication with the vendor's official service endpoints (api.indykite.com). This behavior is documented and represents the primary intent of the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 10:47 AM