Skills
skills/harness/harness-ai/manage-feature-flags/Gen Agent Trust Hub

manage-feature-flags

Pass

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill utilizes a set of predefined MCP tools (harness_list, harness_get, harness_create, harness_execute, harness_update, harness_delete) to interact with the Harness FME (Split.io) backend.
  • [SAFE]: No hardcoded credentials, API keys, or secrets are present. The skill correctly uses placeholders for sensitive identifiers like workspace_id and environment_id.
  • [SAFE]: Analysis of the instructions revealed no prompt injection attempts, obfuscation, or hidden commands.
  • [SAFE]: There are no external downloads or remote code execution patterns. All operations are performed through the specified MCP server (harness-mcp-v2).
  • [SAFE]: The skill's functionality is consistent with its stated purpose of managing feature flag lifecycles.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 15, 2026, 10:16 AM
Security Audit — agent-trust-hub — manage-feature-flags