abilities-api-implement
Pass
Audited by Gen Agent Trust Hub on Jul 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill guides the agent to perform routine development tasks using standard CLI tools, such as the GitHub CLI (
gh pr create) for pull request management and the WordPress CLI (wp eval) for environment inspection and metadata verification. - [EXTERNAL_DOWNLOADS]: The skill references upstream resources and verification tools hosted within the official 'WordPress' GitHub organization (specifically
WordPress/agent-skills). These references target well-known, reputable services and do not involve untrusted remote code. - [PROMPT_INJECTION]: The skill contains logic for processing external audit documents, which represents a potential surface for indirect prompt injection.
- Ingestion points: The agent ingests data from Markdown-based audit documents (
*abilities-audit*.md) located in the project'splans/directory or provided as a runtime argument. - Boundary markers: The process includes mandatory schema validation and explicit human-in-the-loop checkpoints, requiring user confirmation before scaffolding code or opening pull requests.
- Capability inventory: The agent is authorized to create PHP source files, perform git commits, execute shell commands via
wp-env, and interact with the GitHub API via CLI tools. - Sanitization: Input data is validated against a structural schema, and the skill performs cross-verification of capability strings against the local codebase to ensure consistency before use.
Audit Metadata