rc-board

Manages Linear or Jira boards through official MCP servers as a Product Manager.

19|1|Updated Jun 27, 2026
One-click install
npx skills add https://github.com/rodolfochicone/rc-project --skill rc-board-rodolfochicone
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rc-board
Source: https://github.com/rodolfochicone/rc-project/tree/main/skills/misc/rc-board
Command: npx skills add https://github.com/rodolfochicone/rc-project --skill rc-board-rodolfochicone

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Working a product backlog across Linear or Jira requires constant context switching between the tracker, specs, and code. This Skill drives the board through its official MCP server with Product Manager judgment, covering issue creation, refinement into PRD/TechSpec child tasks, and execution with test evidence written back. ## Core Features & Use Cases - Board operations via official MCP: Discuss, create, update, finalize, read, and search issues on Linear or Jira, with provider auto-detection and confirmation gates on every outward-facing write. - Refine an issue into executable work: Turn one issue into a PRD, TechSpec, and native child issues (sub-tasks/sub-issues), recording the board keys in local task files under .rc/tasks/<slug>/. - Execute with evidence: Run child tasks through the implementation flow, persist results to a local sync file first, then post test evidence comments and state transitions to the board, with offline and pending-sync support. - Use Case: A PM asks to refine Jira issue PROJ-123. The Skill generates the PRD and TechSpec, creates sub-tasks as native children, executes them with the test gate, and posts pass/fail evidence back to each child issue. ## Quick Start Ask the agent to create a Jira issue for the new export feature with acceptance criteria and a DoR section, then confirm the preview before it is created.

Frequently Asked Questions about rc-board

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a Jira issue from an AI agent using MCP?▼

Connect the official Atlassian MCP server, then ask the agent to create the issue. It discovers the project, issue types, and required fields via metadata tools, drafts a structured description with acceptance criteria and a DoR section, and creates it only after you confirm the preview.

How do I refine a Linear issue into tasks with an AI agent?▼

Point the agent at the issue and choose the refine operation. It generates a PRD and TechSpec, breaks the work into task files, creates each as a native sub-issue under the parent, and records each sub-issue identifier in the task file frontmatter for later execution.

Does this work with both Linear and Jira boards?▼

Yes, it supports Linear through the official Linear MCP server and Jira through the official Atlassian MCP server. It probes connectivity to detect which provider is configured and asks which board to use when both are reachable.

Can the agent post to my board without my approval?▼

No. Creating issues, adding comments, and moving workflow states are treated as outward-facing writes that require explicit confirmation per action. Reading, searching, and discussing issues need no confirmation.

What happens if the board is unreachable during task execution?▼

Execution results are always written to a local sync file first, so no evidence is lost. When the board becomes reachable again, re-entering the execute flow drains the pending rows and posts them without re-running the tasks.

When should I not use this board workflow?▼

Do not use it for GitHub Issues, bulk imports, or when no board MCP server is configured. It only works through the official Linear or Atlassian MCP servers and never calls provider REST or GraphQL APIs directly.