The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted external data such as user names, emails, and tag descriptions which are then used in command arguments for user creation and updates.
Ingestion points: User information (first name, last name, email, department, role, tags) is processed from the workspace or user input in SKILL.md.
Boundary markers: None detected. The skill lacks explicit instructions to ignore embedded instructions in the processed data.
Capability inventory: The skill can create, update, and delete users, upload files (avatars), and send messages via the Pachca API using the CLI.
Sanitization: No evidence of sanitization or validation of the input strings before they are interpolated into shell commands.
[COMMAND_EXECUTION]: The skill executes shell commands using the vendor's command-line interface tool (pachca and @pachca/cli).
Evidence: Multiple workflows utilize npx -y @pachca/cli and pachca to perform administrative tasks, including user and tag management.
[EXTERNAL_DOWNLOADS]: The skill downloads and installs the official CLI package from the public registry.
Evidence: Use of npx -y @pachca/cli and suggestions to run npm install -g @pachca/cli fetch executable content from the vendor's package repository.

pachca-users