dev_workflow

Project rules for managing software tasks with Task Master, a system for breaking work into tracked tasks and subtasks.

In plain words
What is it for?
They guide task creation, task breakdown, progress management, and restarting the MCP server when its code or tool definitions change.
Why use it?
They define when to use Task Master’s MCP server or command-line tool and how to organize work according to its complexity.

Cursor rule for Cursor

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add rules/learnwithcc/tally-mcp/dev_workflow
Clone the repo
git clone --depth 1 https://github.com/learnwithcc/tally-mcp

Made for: Cursor.

Per session 4,302 This file is loaded in full into every session.
When invoked 4,302 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin 100% copy Near-identical to another mod in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.04302 $0.04302
Opus 5 $0.02151 $0.02151
Sonnet 5 $0.00860 $0.00860
Haiku 4.5 $0.00430 $0.00430

Measured yesterday against content hash de4c7e05df36, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

dev_workflow scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

This is a copy

100% identical to dev_workflow — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

.cursor/rules/dev_workflow.mdc · 239 lines

How it starts

The opening of the file, as written. The whole thing — 239 lines — stays where its author put it; the contents beside it link to each section on GitHub.


description: Guide for using Task Master to manage task-driven development workflows globs: */ alwaysApply: true

Task Master Development Workflow

This guide outlines the typical process for using Task Master to manage software development projects.

Primary Interaction: MCP Server vs. CLI

Task Master offers two primary ways to interact:

  1. MCP Server (Recommended for Integrated Tools):

    • For AI agents and integrated development environments (like Cursor), interacting via the MCP server is the preferred method.
    • The MCP server exposes Task Master functionality through a set of tools (e.g., get_tasks, add_subtask).
    • This method offers better performance, structured data exchange, and richer error handling compared to CLI parsing.
    • Refer to mcp.mdc for details on the MCP architecture and available tools.
    • A comprehensive list and description of MCP tools and their corresponding CLI commands can be found in taskmaster.mdc.
    • Restart the MCP server if core logic in scripts/modules or MCP tool/direct function definitions change.
  2. task-master CLI (For Users & Fallback):

    • The global task-master command provides a user-friendly interface for direct terminal interaction.
    • It can also serve as a fallback if the MCP server is inaccessible or a specific function isn't exposed via MCP.
    • Install globally with npm install -g task-master-ai or use locally via npx task-master-ai ....
    • The CLI commands often mirror the MCP tools (e.g., task-master list corresponds to get_tasks).
    • Refer to taskmaster.mdc for a detailed command reference.

Standard Development Workflow Process

  • Start new projects by running initialize_project tool / task-master init or parse_prd / task-master parse-prd --input='<prd-file.txt>' (see taskmaster.mdc) to generate initial tasks.json
  • Begin coding sessions with get_tasks / task-master list (see taskmaster.mdc) to see current tasks, status, and IDs
  • Determine the next task to work on using next_task / task-master next (see taskmaster.mdc).
  • Analyze task complexity with analyze_project_complexity / task-master analyze-complexity --research (see taskmaster.mdc) before breaking down tasks
  • Review complexity report using complexity_report / task-master complexity-report (see taskmaster.mdc).
  • Select tasks based on dependencies (all marked 'done'), priority level, and ID order
  • Clarify tasks by checking task files in tasks/ directory or asking for user input
  • View specific task details using get_task / task-master show <id> (see taskmaster.mdc) to understand implementation requirements
  • Break down complex tasks using expand_task / task-master expand --id=<id> --force --research (see taskmaster.mdc) with appropriate flags like --force (to replace existing subtasks) and --research.
  • Clear existing subtasks if needed using clear_subtasks / task-master clear-subtasks --id=<id> (see taskmaster.mdc) before regenerating
  • Implement code following task details, dependencies, and project standards
  • Verify tasks according to test strategies before marking as complete (See tests.mdc)
  • Mark completed tasks with set_task_status / task-master set-status --id=<id> --status=done (see taskmaster.mdc)
  • Update dependent tasks when implementation differs from original plan using update / task-master update --from=<id> --prompt="..." or update_task / task-master update-task --id=<id> --prompt="..." (see taskmaster.mdc)
  • Add new tasks discovered during implementation using add_task / task-master add-task --prompt="..." --research (see taskmaster.mdc).
  • Add new subtasks as needed using add_subtask / task-master add-subtask --parent=<id> --title="..." (see taskmaster.mdc).
  • Append notes or details to subtasks using update_subtask / task-master update-subtask --id=<subtaskId> --prompt='Add implementation notes here...\nMore details...' (see taskmaster.mdc).
  • Generate task files with generate / task-master generate (see taskmaster.mdc) after updating tasks.json
  • Maintain valid dependency structure with add_dependency/remove_dependency tools or task-master add-dependency/remove-dependency commands, validate_dependencies / task-master validate-dependencies, and fix_dependencies / task-master fix-dependencies (see taskmaster.mdc) when needed
  • Respect dependency chains and task priorities when selecting work
  • Report progress regularly using get_tasks / task-master list
  • Reorganize tasks as needed using move_task / task-master move --from=<id> --to=<id> (see taskmaster.mdc) to change task hierarchy or ordering

Read the full file on GitHub · 239 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 239 lines · 4,302 tokens per session scan A de4c7e05df36

Subscribe to this mod's changes

dev_workflow is a cursor rule published in the GitHub repository learnwithcc/tally-mcp (8 stars, last pushed 1y ago), licensed ISC. It adds 4,302 tokens to every session, about $0.0215 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to dev_workflow, differing in 0 lines, and is treated as a copy.