The Agent Skills Directory

[SAFE]: The skill operates entirely on local files, performing data transformation from concept YAMLs into strategy draft YAMLs. It does not use network operations or execute external commands.
[DATA_EXPOSURE]: The script reads from a user-specified path and writes to a user-specified output directory. It uses yaml.safe_load for parsing, which prevents common YAML-based injection attacks. No hardcoded credentials or sensitive file access patterns were identified.
[REMOTE_CODE_EXECUTION]: There are no instances of eval(), exec(), or subprocess calls that would allow for arbitrary code execution. All logic is contained within standard Python data processing routines.
[INDIRECT_PROMPT_INJECTION]: While the skill processes external data (edge_concepts.yaml), its output is restricted to structured YAML/JSON files. It lacks the capabilities (like network access or shell execution) required to weaponize malicious input. It uses sanitize_identifier to ensure generated IDs are safe.

edge-strategy-designer