The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on executing several command-line utilities, including lim, npm, make, and openssl, to perform builds, run tests, and manage certificates.
[EXTERNAL_DOWNLOADS]: The skill requires installing the lim package globally using NPM (npm install --global lim). This tool is essential for syncing source code to the remote build environment.
[CREDENTIALS_UNSAFE]: The skill instructions involve handling sensitive developer secrets such as P12 certificates, P12 passwords, and App Store Connect API keys. Although it recommends using environment variables, the agent is directly involved in managing these sensitive credentials.
[INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection because the agent is instructed to read, interpret, and resolve errors found in project files (like project.yml) and build logs. A malicious repository could embed instructions to manipulate the agent's behavior.
Ingestion points: Reads project configuration files and build stdout/stderr logs.
Boundary markers: None identified; the instructions do not tell the agent to ignore potentially malicious embedded content.
Capability inventory: Shell command execution, file system access, and uploading of artifacts to remote servers.
Sanitization: No validation or sanitization of the content from ingested files is mentioned.

limrun-xcode