skills/biw/skills/conductor-setup/Gen Agent Trust Hub

conductor-setup

Pass

Audited by Gen Agent Trust Hub on Jul 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill creates and executes shell scripts such as conductor-setup.sh and conductor-run.sh to manage development environments. It explicitly instructs the agent to perform local checks by executing the scripts it modifies to verify configuration changes.- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from repository configuration files and uses that data to generate or update executable shell scripts.\n
  • Ingestion points: .conductor/settings.toml, conductor.json, and .worktreeinclude files within the repository.\n
  • Boundary markers: The skill lacks explicit instructions to treat configuration values as untrusted data or to use delimiters when interpolating migrated settings into new scripts.\n
  • Capability inventory: The skill has the ability to write to the file system and execute arbitrary shell commands via the scripts.setup and scripts.run workflows.\n
  • Sanitization: There is no mention of sanitizing or validating the content of legacy configuration fields before they are migrated into new executable settings.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 29, 2026, 06:37 PM
Security Audit — agent-trust-hub — conductor-setup