iap-finalizer

Pass

Audited by Gen Agent Trust Hub on Aug 2, 2026

Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection by processing data from untrusted project files.
  • Ingestion points: The agent is instructed to read metadata from .planning/MONETIZATION.md and .planning/PLAN.md.
  • Boundary markers: Absent; there are no instructions to use delimiters or ignore embedded commands.
  • Capability inventory: The skill utilizes Bash to execute a local Python script and uses MCP tools for interacting with the App Store Connect API.
  • Sanitization: Absent; the skill does not explicitly validate or sanitize data extracted from the planning files before API use.
  • [DATA_EXFILTRATION]: The skill documentation references sensitive credential storage locations.
  • Prerequisite instructions mention ~/.appstoreconnect/ for storing API keys, which is a sensitive directory that could be targeted for exposure.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 2, 2026, 01:18 PM
Security Audit — agent-trust-hub — iap-finalizer