Skills
skills/launchdarkly/agent-skills/aiconfig-custom-metrics/Gen Agent Trust Hub

aiconfig-custom-metrics

Pass

Audited by Gen Agent Trust Hub on Apr 22, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements lifecycle management for LaunchDarkly business metrics using the official API at app.launchdarkly.com and the verifiable ldclient-py SDK. All network operations are directed to the service's legitimate infrastructure.
  • [SAFE]: The get_launchdarkly_api_key function in SKILL.md includes logic to auto-detect the LAUNCHDARKLY_API_KEY from environment variables or the Claude MCP configuration file (~/.claude/config.json). This is a documented integration helper that specifically targets the vendor's own configuration key.
  • [SAFE]: No indicators of malicious behavior such as prompt injection, obfuscation, persistence mechanisms, or unauthorized privilege escalation were identified.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 22, 2026, 07:12 PM