The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a local CLI tool (workflow) and an underlying Python script (~/.codex/bin/workflow.py) for tasks such as system diagnostics, documentation retrieval, and linting.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates the reading and processing of external documentation from ~/.codex/docs/, which creates a standard surface for indirect prompt injection if those files contain malicious instructions.
Ingestion points: Documentation files accessed via workflow read or the Read tool.
Boundary markers: Not present; the instructions do not specify delimiters to separate document content from agent instructions.
Capability inventory: Includes file system read access (Read, Glob, Grep) and file modification capabilities via the workflow add and workflow update commands.
Sanitization: No explicit sanitization or validation of document content is mentioned.

workflow-cli