skills/fradser/skills/start-feature/Gen Agent Trust Hub

start-feature

Pass

Audited by Gen Agent Trust Hub on Aug 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a bundled shell script scripts/start-branch.sh and multiple git CLI commands to manage the repository state.
  • [DATA_EXFILTRATION]: The skill uses git push to synchronize local changes with remote repositories.
  • [PROMPT_INJECTION]: The skill identifies potential indirect prompt injection surfaces where external data is ingested.
  • Ingestion points: User-provided $ARGUMENTS and commit history via git log.
  • Boundary markers: None explicitly defined for commit messages.
  • Capability inventory: Shell script execution, file modification (sed), and git management.
  • Sanitization: Branch names are slugified and versions are validated against semver patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 18, 2026, 03:52 PM
Security Audit — agent-trust-hub — start-feature