The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes local shell commands and Node.js for project triage and code searching. Specifically, it calls node skills/wp-project-triage/scripts/detect_wp_project.mjs to identify WordPress environments. This is a standard procedure for a developer-focused agent skill.
[PROMPT_INJECTION]: No evidence of direct prompt injection or attempts to override AI safety filters was found. The instructions are strictly limited to WordPress development procedures.
[EXTERNAL_DOWNLOADS]: There are no remote code downloads, external scripts, or network requests initiated by the provided skill files.
[CREDENTIALS_UNSAFE]: No hardcoded secrets, API keys, or credentials were found in the skill or its references.
[DATA_EXFILTRATION]: The skill does not perform any network operations that could lead to data exfiltration. It operates entirely on the local repository root.
[INDIRECT_PROMPT_INJECTION]: The skill has a low risk of indirect prompt injection as it processes untrusted local codebases.
Ingestion points: Reads files and project structure in the repository root.
Boundary markers: None explicitly defined for the triage or search steps.
Capability inventory: Uses bash, node, and potentially WP-CLI via subprocess calls.
Sanitization: Relies on the agent's internal filtering when processing search results from grep or similar tools.

wp-rest-api