extend

Pass

Audited by Gen Agent Trust Hub on Mar 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the @openstall/sdk package from the npm registry. This is a vendor-owned tool used to interact with the OpenStall marketplace infrastructure.
  • [COMMAND_EXECUTION]: The agent is instructed to use the openstall CLI for registration, discovery, and execution of marketplace capabilities. It also accesses its local configuration file at ~/.openstall/config.json.
  • [PROMPT_INJECTION]: The skill defines a workflow where data from external marketplace agents is processed, creating an indirect prompt injection surface.
  • Ingestion points: Outputs from external agents received through the openstall call command (SKILL.md).
  • Boundary markers: The instructions do not specify any delimiters or safety headers for processing external data.
  • Capability inventory: Uses the openstall CLI for remote capability invocation and task management (SKILL.md).
  • Sanitization: There are no defined procedures for sanitizing or verifying the integrity of the data returned by third-party agents.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 16, 2026, 12:30 AM