insforge-cli

Pass

Audited by Gen Agent Trust Hub on Oct 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the InsForge CLI (@insforge/cli) extensively to manage cloud infrastructure, including database migrations, storage, functions, and compute services. All commands are executed via npx to ensure the correct version and are routed through authenticated platform endpoints.
  • [INDIRECT_PROMPT_INJECTION]: The skill exposes an attack surface for indirect prompt injection through commands that ingest untrusted data, such as memory recall, diagnose --ai, and backend log inspection.
  • Ingestion points: npx -y @insforge/cli memory recall, npx -y @insforge/cli diagnose --ai, and npx -y @insforge/cli logs (diagnostics.md).
  • Boundary markers: The skill encourages the use of the --json flag for structured output to assist programmatic parsing.
  • Capability inventory: The skill has broad capabilities to modify database schema, deploy code, and manage secrets via the CLI (SKILL.md).
  • Sanitization: Mentions automatic redaction of common sensitive patterns (emails, credentials) in the feedback command reporting flow.
  • [DYNAMIC_EXECUTION]: Supports the deployment of Deno-based edge functions and the creation of PostgreSQL SECURITY DEFINER functions. The provided documentation includes high-quality security warnings regarding search_path pinning and recursion risks in database access control (access-control.md).
  • [EXTERNAL_DOWNLOADS]: Fetches required CLI tools and wizards from well-known registries or official installers. Specifically, it uses @insforge/cli and @posthog/wizard from NPM, and mentions the official Fly.io installer.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 24, 2026, 09:11 PM