launchdarkly-release-autopilot-settings
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources, including GitHub pull request metadata and repository configurations retrieved via MCP tools. This creates a potential surface for indirect prompt injection if an attacker embeds instructions in PR titles or descriptions. The skill mitigates this by enforcing strict human-in-the-loop (HITL) confirmation policies ('Confirm before write') that require the agent to display the current versus proposed state and wait for an explicit user 'yes' before executing any state-modifying operations (PATCH, map, unmap).
Audit Metadata