Loading ToolBench
Bridges OpenSpec change management with Claude Flow multi-agent orchestration via Claude Code
OpenSpec-Flow has 12 tools with complete input schemas and descriptions, but quality is uneven. Naming follows verb_noun convention well (get_, list_, create_, analyze_, split_, archive_). Descriptions are present (avg ~120 chars) but lack actionable context, many omit WHEN to use the tool or what happens on failure. Parameter descriptions are minimal; many lack format/constraint guidance. Output schemas are undocumented, LLMs cannot predict response structure. Error handling is absent from descriptions. The archive_change tool is marked IRREVERSIBLE but has no confirmation pattern or dry-run option. Overall: functional but below production baseline for agent-facing tools.
Analyze an OpenSpec change for size, complexity, and splitting recommendation
Analyze incomplete tasks for an OpenSpec change and generate a deferred items report
Archive a completed or closed OpenSpec change using the OpenSpec CLI. Requires openspec CLI to be installed.
Create an implementation flow log for an OpenSpec change
Generate a comprehensive work brief for an OpenSpec change
Get context for an OpenSpec change: file paths, summary, and config. Agents should Read files as needed.
Get the OpenSpec proposal workflow instructions. Reads from .claude/commands/openspec/proposal.md if it exists, providing the canonical workflow for creating OpenSpec changes.
No output schemas documented. LLMs cannot predict response structure, forcing them to guess what fields are returned and how to chain tools. E.g., list_changes returns what? Array of objects with which fields?
archive_change is marked IRREVERSIBLE but lacks confirmation/dry-run pattern. No description warns of consequences. Agents may accidentally archive changes without understanding the action is permanent.
Parameter descriptions lack format/constraint guidance. E.g., change_id is described as 'The OpenSpec change ID' but no format is specified (alphanumeric? slug? UUID?). LLMs cannot validate inputs without explicit constraints.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 53 | 2026-07-28+ | v2 |
List all OpenSpec changes with their status and task completion
Resolve a partial change ID to the full ID. Returns exact match, resolved match, ambiguous matches, or not found with suggestions.
Save or update an artifact file for an OpenSpec change
Create a new OpenSpec change directory structure with initialized files
Split a large OpenSpec change into phased sub-changes
Descriptions lack WHEN-to-use context. E.g., 'Analyze incomplete tasks for an OpenSpec change' does not explain when an agent should call this vs. get_change_context. No recovery guidance for errors.
No error handling guidance in tool descriptions. What happens if change_id is not found? What should the agent do next? Descriptions must include recovery paths.