KS
Killer-Skills

brain-dump-workflow — how to use brain-dump-workflow how to use brain-dump-workflow, brain-dump-workflow setup guide, what is brain-dump-workflow, brain-dump-workflow alternative, brain-dump-workflow vs competitor, brain-dump-workflow install, MCP tools tutorial, AI-assisted development workflow, universal quality workflow

v1.0.0
GitHub

About this Skill

Perfect for Development Agents needing structured workflow management with MCP tools and a 5-step sequence. brain-dump-workflow is a universal quality workflow designed from the ground up with and for AI-assisted development, featuring a canonical source in src/api/ralph.ts

Features

Provides a 5-step sequence using MCP tools
Consolidates 9 tools with action-dispatched parameters
Requires passing 'action' as the first parameter to specify operations
Utilizes the workflow tool with 'a' as the starting point
Includes a system prompt with the full workflow
References the canonical source in src/api/ralph.ts

# Core Topics

salmanrrana salmanrrana
[0]
[0]
Updated: 3/6/2026

Quality Score

Top 5%
42
Excellent
Based on code quality & docs
Installation
SYS Universal Install (Auto-Detect)
Cursor IDE Windsurf IDE VS Code IDE
> npx killer-skills add salmanrrana/brain-dump/brain-dump-workflow

Agent Capability Analysis

The brain-dump-workflow MCP Server by salmanrrana is an open-source Categories.community integration for Claude and other AI agents, enabling seamless task automation and capability expansion. Optimized for how to use brain-dump-workflow, brain-dump-workflow setup guide, what is brain-dump-workflow.

Ideal Agent Persona

Perfect for Development Agents needing structured workflow management with MCP tools and a 5-step sequence.

Core Value

Empowers agents to execute a provider-agnostic universal quality workflow, utilizing 9 consolidated MCP tools with action-dispatched parameters, and enforcing a mandatory 5-step sequence for standardized development processes, leveraging the `getRalphPrompt()` function from `src/api/ralph.ts`.

Capabilities Granted for brain-dump-workflow MCP Server

Initializing projects with the workflow tool
Dispatching actions with specified parameters
Enforcing the mandatory 5-step sequence for consistent development

! Prerequisites & Limits

  • Requires adherence to the 5-step sequence
  • MCP tools must be used with action-dispatched parameters
Project
SKILL.md
3.6 KB
.cursorrules
1.2 KB
package.json
240 B
Ready
UTF-8

# Tags

[No tags]
SKILL.md
Readonly

Brain Dump Universal Quality Workflow

Canonical source: src/api/ralph.ts -> getRalphPrompt(). This skill provides a quick reference. The system prompt has the full workflow.

MCP tools are consolidated: 9 tools with action-dispatched params. Always pass action as the first parameter to specify the operation.

MANDATORY 5-Step Sequence

Every ticket MUST go through these steps using MCP tools. Never skip any.

Step 1: Start Work

Call the workflow tool with action: "start-work" and ticketId: "<id>" BEFORE writing any code.

This creates a git branch, sets status to in_progress, and posts a "Starting work" comment.

Step 2: Implement + Verify

Write code, then run quality gates:

  • pnpm type-check -- must pass
  • pnpm lint -- must pass
  • pnpm test -- must pass

Commit with format: feat(<ticket-short-id>): <description>

Step 3: Complete Implementation

Call the workflow tool with action: "complete-work", ticketId: "<id>", and summary: "<what you did>".

This moves ticket to ai_review and posts a work summary comment.

Step 4: AI Review (via MCP tools -- NOT local review skills)

IMPORTANT: Do NOT use local /review skills, subagents, or code review tools. Perform self-review by reading your own diffs, then record findings via MCP.

For each issue found, call the review tool:

action: "submit-finding"
ticketId: "<ticket-id>"
agent: "code-reviewer"       // or "silent-failure-hunter" or "code-simplifier"
severity: "major"            // critical | major | minor | suggestion
category: "error-handling"
description: "Clear description of the issue"

Fix critical/major issues, then call the review tool:

action: "mark-fixed"
findingId: "<finding-id>"
fixStatus: "fixed"

Verify by calling the review tool with action: "check-complete" and ticketId: "<id>" -- response must contain canProceedToHumanReview: true.

Step 5: Generate Demo + STOP

Call the review tool:

action: "generate-demo"
ticketId: "<ticket-id>"
steps: [{ order: 1, description: "...", expectedOutcome: "...", type: "manual" }]

Include 3-7 manual test steps. Ticket moves to human_review.

STOP HERE. Do NOT continue. Only humans can approve tickets.

DO NOT

  • Skip any step above
  • Set ticket status to "done" directly
  • Continue working after generating demo
  • Write code before calling workflow start-work
  • Use local review skills or subagents instead of review submit-finding
  • Describe demo steps in text instead of calling review generate-demo
  • Create git branches manually instead of using workflow start-work

Severity Guide (for Step 4)

SeverityWhen to Use
criticalBug that breaks functionality or causes crashes
majorIncorrect behavior or error handling issue
minorCode quality issue, not a bug
suggestionNice-to-have improvement

Demo Step Types (for Step 5)

TypeWhen to Use
manualUser performs an action
visualUser visually confirms something
automatedSystem runs a command/test

Reference Docs

For detailed guidance on specific phases, see:

  • reference/review-guide.md -- Self-review checklist and review agent details
  • reference/troubleshooting.md -- Common errors and recovery steps
  • reference/git-linking.md -- Linking commits and PRs to tickets
  • reference/compliance-logging.md -- Enterprise conversation logging

Related Skills

Looking for an alternative to brain-dump-workflow or building a Categories.community AI Agent? Explore these related open-source MCP Servers.

View All

widget-generator

Logo of f
f

widget-generator is an open-source AI agent skill for creating widget plugins that are injected into prompt feeds on prompts.chat. It supports two rendering modes: standard prompt widgets using default PromptCard styling and custom render widgets built as full React components.

149.6k
0
Design

chat-sdk

Logo of lobehub
lobehub

chat-sdk is a unified TypeScript SDK for building chat bots across multiple platforms, providing a single interface for deploying bot logic.

73.0k
0
Communication

zustand

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
Communication

data-fetching

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
Communication