The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill directs users to download an installation script from the author's GitHub repository (github.com/yigitkonur/codex-worker) and provides a fallback installation method via the npm package registry.
[REMOTE_CODE_EXECUTION]: Installation steps include piping a remote shell script directly to bash or sudo bash, allowing for the execution of code hosted on a remote server.
[COMMAND_EXECUTION]: Instructions involve using sudo to obtain administrative permissions for global tool installation and executing various codex-worker subcommands to manage threads and turns.
[PROMPT_INJECTION]: The skill is designed to ingest and act upon content from markdown files (e.g., task.md), which constitutes an indirect prompt injection surface where untrusted data could influence the agent's actions. * Ingestion points: Markdown prompt files are read and processed by the CLI as described in SKILL.md and references/command-reference.md. * Boundary markers: There are no explicit markers or instructions to isolate or ignore potentially malicious content within the ingested files. * Capability inventory: The codex-worker CLI can modify files and execute commands within the workspace (documented in references/orchestration-patterns.md). * Sanitization: No sanitization or verification of the markdown file content is specified before execution.

run-codex-bridge