The Agent Skills Directory

[DATA_EXPOSURE] (SAFE): The skill reads project source files, including profiles and configuration files, to synchronize documentation. This access is local and inherent to the skill's purpose. No patterns of data exfiltration or unauthorized access to sensitive user data (like SSH keys or AWS credentials) were found.
[COMMAND_EXECUTION] (SAFE): The skill's instructions guide the agent to perform file I/O operations and consistency checks within the project directory. There is no evidence of arbitrary command execution, shell injection, or unauthorized system modifications.
[PROMPT_INJECTION] (SAFE): The instructions are focused on documentation workflows. There are no attempts to bypass safety filters, extract system prompts, or override agent constraints.
[NO_CODE] (SAFE): The skill consists solely of Markdown documentation and reference files. It does not include or execute any scripts, binaries, or external dependencies, which significantly limits the attack surface.

synapse-docs