Loading ToolBench
Time tracking system with agent session support, CLI interface, and Clockify integration
BurrowTime defines 18 tools with basic schemas and descriptions, but falls short of production quality. Most tools have descriptions (good), but they are terse (avg ~60 chars, well below the 194-char baseline). Parameter descriptions are sparse or missing entirely. No output schemas are documented. Error handling is absent, tools do not guide recovery or categorize failures. Tool names follow verb_noun convention (agent start, clockify sync), but descriptions lack context on WHEN to use each tool or what happens on failure. No tool annotations (readOnlyHint, destructiveHint, idempotentHint) despite many WRITE operations. Schemas are present but minimal, most parameters lack type constraints, enums, or validation guidance. The 'agent' tool family is well-structured (start, heartbeat, pause, resume, stop, status, recover, report), but the 'clockify' family mixes configuration, querying, and sync operations without clear separation of concerns.
Renew an active agent session lease.
Pause an agent session and close its current standard timer.
Stop agent timers whose leases expired.
Report time grouped by agent session.
Resume a paused agent session with a new standard timer.
Start an agent-owned session using a standard concurrent timer.
List agent-owned sessions.
Stop an agent session without affecting other timers.
No output schemas documented. Tools return results but LLMs cannot predict field names, types, or structure. Downstream tool chaining is fragile.
Descriptions are terse (avg ~50 chars vs 194-char baseline). Most lack context on WHEN to use the tool, what happens on failure, or prerequisites. E.g., 'Renew an active agent session lease' does not explain what happens if the lease expires or how to handle renewal failures.
Many parameters lack descriptions or type constraints. E.g., 'tags' in agent start is an array but no description explains the format, allowed values, or max count. 'rounding' in clockify configure has no enum constraint despite being a fixed set (off, up, nearest).
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 43 | 2026-07-28+ | v2 |
Show machine-readable BurrowTime features.
Save a connection without storing the API key
Route a local project to this connection
List accessible Clockify project IDs
Verify an existing remote entry and complete a pending receipt
Allow retry only after verifying an uncertain upload did not create an entry
Show local export receipts, including pending uploads
Upload completed entries (today by default), skipping recorded uploads
Get and set configuration options.
Rename a project or tag.
No error handling guidance. Tools do not indicate which errors are retryable, user-fixable, or fatal. No recovery hints (e.g., 'If session not found, call agent status to list active sessions'). Agents cannot self-correct on failure.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint). Many tools are WRITE operations (agent start, agent stop, clockify sync, rename, config) but lack destructiveHint. Agents cannot reason about side effects or retry safety.