The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains no instructions designed to bypass safety filters, extract system prompts, or override agent constraints. Its logic is focused on document formatting and mode detection (Create/Update/Find).
[DATA_EXFILTRATION]: No network operations (e.g., curl, wget) or commands targeting sensitive files (e.g., .ssh, .env, .aws) are present. The skill only interacts with standard Markdown templates and schema files located in relative local directories.
[REMOTE_CODE_EXECUTION]: The skill does not perform any remote package installation or script execution. It relies on internal instructions and pre-defined local Markdown structures.
[OBFUSCATION]: No obfuscation techniques such as Base64 encoding, zero-width characters, or homoglyph substitutions were detected. The content is fully transparent and human-readable.
[INDIRECT_PROMPT_INJECTION]: The skill has a defined surface for processing external data through the @file syntax and user-supplied decision types. However, its capabilities are restricted to document generation and saving files to a specific 'charters/' directory, which does not present a mechanism for downstream exploitation or privilege escalation.

decision-charter