The Agent Skills Directory

[SAFE]: The skill instructs the installation of the paperclipai CLI and the use of the @bbengamin/paperclip-mcp-server Node.js package. These tools are provided by the vendor for system interaction and are installed via standard package registries.
[SAFE]: All system mutations, including issue updates and commenting, are governed by a human-in-the-loop approval process, preventing unauthorized autonomous actions or modifications to the organizational state.
[SAFE]: The skill reads external data from wiki pages and issue descriptions. The risk of indirect prompt injection is mitigated by the requirement for human verification of all triaged outcomes and recommended changes.
[SAFE]: API credentials and bearer tokens are managed securely through environment variables, with explicit instructions for the operator to avoid logging or echoing sensitive information during configuration.

outbound-triage