The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown documentation providing conceptual advice for creative writing. It does not include any scripts, commands, or configuration that could pose a security risk.
[INDIRECT_PROMPT_INJECTION]: The skill mentions recording audits in a local file (tension/TENSION.md). While this constitutes a data ingestion surface, the skill lacks any risky capabilities (such as network access, command execution, or file system modifications outside its scope) that could be leveraged for an attack. The instructions are focused on narrative analysis and do not provide a vector for exploitation.

tension-engine