Loading ToolBench
Model Context Protocol server for Jira and Confluence with SSE transport
Server exposes 15 tools with decent naming conventions (verb_noun pattern) and present descriptions. However, multiple critical gaps reduce quality: output schemas are completely undocumented (no return type specifications visible in code), parameter descriptions exist but are generic and lack constraint details, error handling is minimal with no recovery guidance, and tool annotations are absent. Tools are well-named but parameter validation rules and format constraints are not documented in descriptions. Average description length is acceptable (~100-150 chars), but they lack the actionable detail required for LLM optimization (dependencies, prerequisites, when to use vs similar tools).
Add a comment to a Confluence page
Add a comment to a Jira issue
Create a new Confluence page in the specified space
Create a new Jira issue in the specified project
Extract and list all Confluence page links from a Jira issue
Get the full content of a Confluence page by its ID
Get the full details of a Jira issue including comments
No output schemas documented for any tool. Code shows input parameter schemas but returns/outputs are completely undocumented. LLMs cannot know what fields to expect from responses, breaking downstream tool composition and forcing guesswork about field names.
Parameter descriptions lack constraint documentation. E.g., 'limit' parameters have no min/max bounds, 'status' fields have no enum values, 'query' fields lack format specifications (CQL vs JQL vs free text). LLMs cannot validate inputs before calling.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | F | 47 | <=2025-11-25 | v2 |
| 2026-03-09 | F | 42 | - | v1 |
Get issues assigned to the current user
List all available Jira projects
Search Confluence pages using CQL query
Search Jira issues using JQL query with smart query enhancement
Get a comprehensive summary of a Jira issue including all details, comments, and related links
Transition a Jira issue to a new status
Update an existing Confluence page with new content
Update an existing Jira issue with new values
Descriptions lack actionable context. Missing: when to use this tool vs a similar one (search_confluence vs get_confluence_page), prerequisites (must have space_key first?), dependencies on other tools, what data is returned. Descriptions are 'what it does' but not 'when/why to use it'.
No error handling guidance visible. No recovery paths documented (e.g., 'If page not found, call search_confluence first'). Agents hitting failures have no direction on what to try next.
No tool annotations present (readOnlyHint, destructiveHint, idempotentHint). Agents cannot distinguish safe (read-only) tools from destructive ones without additional logic. create_*, update_*, add_* tools should be marked destructive.
Pagination not implemented. search_confluence and search_jira accept 'limit' but no 'offset', 'cursor', or 'page' parameter visible. Agents cannot retrieve results beyond the first page; large result sets are truncated.
update_jira_issue has overlapping parameters. 'summary' and 'description' are listed as optional in the tool definition, but no guidance on which are required or if some combinations are invalid. Unclear what a minimal valid update looks like.
transition_jira_issue expects 'status' but Jira internally uses transitions/status combinations that may not match user input. No mapping guidance provided; LLMs may pass invalid status values.