coding-discipline
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to execute various shell commands as part of the development lifecycle, including syntax checks, build systems, and the GitHub CLI tool.
- Evidence: Instructs usage of
gh apito check Dependabot, code-scanning, and secret-scanning alerts. - Evidence: Mentions running
TOOLS/Get-GitHubFlags.ps1to check repository flags. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon data from external project files which may contain untrusted content.
- Ingestion points: Reads from
TASK.md,ENVIRONMENT.md, and other local project files. - Boundary markers: None specified in the instructions for isolating external content.
- Capability inventory: The agent is authorized to read/write files and execute shell commands (builds, tests,
gh). - Sanitization: The instructions advocate for the use of typed parsers and structured libraries over regex to mitigate parsing risks.
Audit Metadata