How do I install the Batch Issues skill?

Run npx skillmds@latest add future-architect/batch-issues in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

What does the Batch Issues skill do?

Batch-process open GitHub issues with parallel agents. Analyzes conflicts, dispatches workers, creates PRs. Usage: /batch-issues [issue numbers] [--dry-run] [--max-parallel N] It is listed under Coding & Dev Tools on SkillMD.

Is Batch Issues safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Batch Issues?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Batch Issues free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Batch Issues?

future-architect (@future-architect) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Batch Issues by future-architect — Agent Skill

Name: Batch Issues
Author: future-architect

File contents

/batch-issues -- Parallel Issue Dispatcher

Full specification: See .github/prompts/batch-issues.prompt.md for the complete procedure, conflict analysis, agent workflow, and safety rules.

Quick Reference

Auto-select: /batch-issues -- pick open issues, analyze conflicts, dispatch
Specific issues: /batch-issues 168 171 187 -- process these issues
Dry run: /batch-issues --dry-run -- show plan only
Limit parallelism: /batch-issues --max-parallel 2
Phase 1: Issue selection & filtering
Phase 2: Conflict analysis (shared file detection)
Phase 3: Size/risk assessment (S/M/L classification)
Phase 4: Dispatch plan (present to user)
Phase 5: Agent dispatch (worktree-isolated, parallel across groups, sequential within)
Phase 6: Summary report with PR links
Phase 6.5: Orchestrator review — run /review-until-clean on each PR (agents can't launch independent reviewers)
Phase 7: Evidence verification — check/fix Before/After test output in every PR body
Each agent: reproduce (MANDATORY) -> architect -> implement -> self-check -> verify after (MANDATORY) -> before/after PR (template enforced)
Independent review runs at orchestrator level (Phase 6.5), NOT inside agents

Frequently asked questions

Batch Issues