flyingnobita-deliver-setup

Pass

Audited by Gen Agent Trust Hub on Nov 10, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted repository-level configuration files (such as delivery.toml, repository rules, and AGENTS.md) to determine setup parameters and migration steps.
  • Ingestion points: SKILL.md (Step 1, Step 3) and configuration files referenced during prerequisite checks in references/prerequisites.md.
  • Boundary markers: No explicit instructions or delimiters are provided to the agent to prevent the interpretation of instructions that might be embedded within the configuration data.
  • Capability inventory: The skill can execute shell commands for environment checks, write configuration files, and perform repository commits.
  • Sanitization: There is no evidence of validation or sanitization of ingested configuration values before they are used in logic or displayed to the user.
  • [COMMAND_EXECUTION]: Executes multiple system commands to verify the presence and status of required tools and identities.
  • Evidence: Runs git config, gh auth status, python3 -c 'import tomllib', herdr --version, and ORCA --version (or the command specified in ORCA_CLI_COMMAND).
  • [EXTERNAL_DOWNLOADS]: Provides references and installation commands for third-party tools and agent skills from external sources.
  • Evidence: Points to GitHub repositories for mattpocock/skills, herdrdev/herdr, stablyai/orca, and getpaseo/paseo to resolve missing dependencies.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 10, 2026, 08:50 AM