LayrShift exposes 6 tools with mixed quality. Tool names follow verb_noun convention (create-, delete-, edit-, disable-) which is good. However, parameter descriptions are sparse or missing entirely. Input schemas are present but lack depth, many parameters have no description text. Output schemas are not documented. Error handling guidance is absent. The server targets dev/staging WordPress environments with high-risk operations (delete-file, create-admin-access-link), yet lacks confirmation patterns or dry-run support for destructive actions. Tool descriptions are adequate (50-100 chars) but parameter-level documentation is weak.
Creates a temporary, one-time WordPress admin access exchange for browser automation tools.
Creates a temporary upload link for file uploads.
Deletes a file or directory from the WordPress installation.
Disables a file by moving it to a sandbox disabled directory.
Edits a file by replacing a string with new content.
Discover all available WordPress abilities, plus LayrShift environment instructions.
discover-abilities has no visible input schema or output schema documentation. Tool definitions are inferred from registry, not explicitly visible in source.
delete-file and edit-file lack confirmation/dry-run patterns despite being destructive/write operations. No guidance on recovery if operation fails or is misapplied.
Parameter descriptions are minimal or absent. 'path' in delete-file has no format/constraint guidance. 'old_string' and 'new_string' in edit-file lack examples or length limits. LLMs cannot infer valid input ranges.
No output schemas documented for any tool. Agents cannot plan downstream calls or extract required fields (e.g., what does create-admin-access-link return? token? expiry? redirect URL?).
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 55 | <=2025-11-25 | v2 |
Error handling is not visible in tool definitions. No guidance on retryability, user-fixable errors, or recovery steps. Agents will not know how to respond to failures.