The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses uv run python3 to execute various local scripts such as scope_check.py, brain.py, and global_brain.py. These scripts are used for target validation, state management, and data synchronization.
[DATA_EXFILTRATION]: The command uv run python3 $CLAUDE_PROJECT_DIR/tools/global_brain.py sync-from-local indicates that the skill synchronizes local reconnaissance and scan summaries to a global repository. This is a standard feature for distributed security testing tools.
[PROMPT_INJECTION]: The skill ingests and processes data from external sources during the reconnaissance and scanning phases, which presents an indirect injection surface.
Ingestion points: Data enters the context via the scope.yaml file and the output of the recon, config-auditor, js-analyzer, and vuln-scanner agents.
Boundary markers: The instructions do not define explicit delimiters or 'ignore' instructions for the data processed from these agents.
Capability inventory: The skill has the ability to execute local Python scripts and dispatch further agents based on the processed data.
Sanitization: There is no evidence of explicit sanitization or validation of the results returned by the specialized scanning agents before they are recorded in the 'brain'.

pipeline