create-secret

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user-supplied content such as secret values and descriptions to be processed by tools like harness_create. This creates a vulnerability where malicious inputs could attempt to influence the agent's logic.
  • Ingestion points: User-provided metadata and values within SKILL.md and references/secret-types.md.
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to treat user-supplied content as non-executable or to ignore embedded prompts.
  • Capability inventory: The skill uses tools (harness_create, harness_update, harness_delete) that can change the state of external infrastructure via network calls.
  • Sanitization: No sanitization or validation steps are defined for the user-supplied data before it is passed to the MCP tools.
  • [COMMAND_EXECUTION]: The skill defines operations using a suite of MCP tools (harness_create, harness_update, harness_delete, harness_list, harness_get). These tools allow the agent to execute management commands that affect the Harness environment and its resources.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 11:25 PM
Security Audit — agent-trust-hub — create-secret