Desktop-local agentic engine with real-time chat, multiple LLM providers, and controlled automation. FastAPI backend with WebSocket support, authentication, and multi-agent orchestration.
Open Slap exposes 2 tools for autoapprove management. Both have descriptions and basic input schemas, but critical gaps undermine production readiness. Tool names lack clear verb-noun structure (commands_list_autoapprove vs list_autoapprove). Parameter descriptions are minimal (1-2 sentences). Output schemas are not documented, LLMs cannot infer what fields to expect from list results or confirm deletion success. No error handling guidance. The READ_ONLY and WRITE risk annotations are present but not integrated into tool definitions as formal idempotentHint/destructiveHint annotations per current MCP spec.
Delete a user command autoapprove entry
List user command autoapprove entries with pagination
Tool names do not follow verb_noun convention. 'commands_list_autoapprove' should be 'list_autoapprove' and 'commands_delete_autoapprove' should be 'delete_autoapprove'. The 'commands_' prefix is redundant and obscures intent.
Output schemas are not documented. list_autoapprove returns paginated results but the response structure (fields, types, pagination metadata) is not specified. LLMs cannot plan downstream operations or extract required data.
delete_autoapprove lacks confirmation or dry-run capability. Destructive operations should support a confirmation step to prevent accidental data loss by agents.
Parameter descriptions are minimal (1-2 sentences). 'Normalized command string to delete' does not explain what 'normalized' means, valid format, or how to obtain it. Descriptions should be 50-200 chars and include format/constraint hints.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 45 | 2026-07-28+ | v2 |
No error handling guidance. If delete fails (command not found, permission denied), the tool should return actionable recovery hints, not raw error codes.