The Agent Skills Directory

[COMMAND_EXECUTION]: The skill workflow involves executing a local preflight tool and a Python adapter script to process audit data. The test suite also utilizes subprocess calls to verify script functionality. These executions are scoped to local files and tools required for the skill's primary auditing purpose.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted repository data, including AGENTS.md and application source code. The dash_ui_audit_adapter.py script includes specific logic to redact absolute repository paths and handle malformed input, which helps mitigate the risk of sensitive path exposure or malformed data influencing the agent's behavior during the audit process.

dash-audit