cleaning-up-stale-feature-flags

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted external data retrieved from the PostHog API during its auditing workflow.
  • Ingestion points: The skill ingests flag names, descriptions, and metadata via the posthog:feature-flag-get-all and posthog:feature-flag-get-definition tools as described in the workflow of SKILL.md.
  • Boundary markers: No specific boundary markers or delimiters are defined to isolate or instruct the agent to ignore potentially malicious content within flag metadata.
  • Capability inventory: The skill uses tools capable of modifying the project state, specifically posthog:feature-flag-disable and posthog:delete-feature-flag.
  • Sanitization: The skill does not specify any sanitization or validation of the flag data retrieved from the API.
  • Note: The risk is significantly mitigated by mandatory user confirmation and deployment check steps before any flags are disabled or deleted.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 03:32 PM
Security Audit — agent-trust-hub — cleaning-up-stale-feature-flags