A Spring Boot-based AI agent that uses MCP tools to assist with software engineering tasks, including ticket resolution, code modification, and GitHub integration. Routes user intents to specialized workflows and integrates with LLMs via Spring AI.
This server has severe definition quality issues. Tool descriptions are minimal or in Spanish without English translations. Input schemas are partially visible but lack proper type definitions and parameter descriptions. The safeGithubPush tool has a Spanish description ('ÚSALA SIEMPRE...') that is non-standard for production MCP. Two tools (jira_issues, read_file) have no visible input schemas in the provided code. Output schemas are not documented. Error handling is minimal. No tool annotations (readOnlyHint, destructiveHint, idempotentHint) are present despite two tools being WRITE operations. The server appears to be a proof-of-concept rather than production-grade.
Retrieves Jira issues for ticket resolution workflows
MCP tool for pushing files to GitHub repositories
Reads file contents from the local filesystem
ÚSALA SIEMPRE PARA SUBIR ARCHIVOS A GITHUB. Lee el archivo local y lo sube.
Two tools (jira_issues, read_file) have no visible input schemas in source code.
Tool descriptions are inadequate. safeGithubPush uses Spanish ('ÚSALA SIEMPRE PARA SUBIR ARCHIVOS A GITHUB') instead of English. jira_issues and read_file have generic 20-char descriptions.
No tool annotations present. safeGithubPush and push_files are WRITE operations that should declare destructiveHint=true. read_file should declare readOnlyHint=true. Missing annotations prevent LLMs from understanding operation safety.
Parameter descriptions are minimal or missing. safeGithubPush parameters have basic descriptions but lack format constraints (e.g., 'filePath' should specify absolute path format). No enum constraints or validation guidance.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 34 | 2026-07-28+ | v2 |
No output schemas documented. Tools return strings or objects but LLMs cannot predict response structure.