tigris-security-access-control

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill install s the @tigrisdata/cli package from the npm registry. This is an official vendor tool provided by Tigris Data for managing their storage service s.
  • [COMMAND_EXECUTI O N]: The skill utilize s the tigris CLI to execute storage management command s, including key creation, bucket assignment, and CORS configuration. The se operation s are fundamental to the skill's described purpose.
  • [INDIRECT_PROMPT_INJECTI O N]: The skill ingest s external data such as CORS configuration file s and storage bucket metadata, presenting a potential surface for indirect prompt injection. 1. Ingestion points: The skill read s conten t from cors.json file s and retrieve s metadata from Tigris storage bucket s during audit s (SKILL.md). 2. Boundary markers: No specifi c delim iter s or instruction s are defined to differentiate between external data and agent instruction s. 3. Capability inventory: The skill is capable of executing shell command s via the tigris CLI, which include s privilege s to modify resource s and access policie s. 4. Sanitization: The re are no specifi ed sanitization or validation step s for the conten t of the ingested configuration file s.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 08:17 PM