workflow

A workflow for managing coding tasks, including writing a product requirements document, tracking tasks, researching before edits, and updating project history.

In plain words
What is it for?
It guides non-trivial feature work and bug fixes from planning through testing and changelog updates.
Why use it?
It prevents unplanned implementation and keeps the work, verification, and documentation traceable.

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/waynesutton/markdown-site/workflow
Clone the repo
git clone --depth 1 https://github.com/waynesutton/markdown-site

Made for: Cursor.

Per session 512 This file is loaded in full into every session.
When invoked 512 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found 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 $0.00512 $0.00512
Opus 5 $0.00256 $0.00256
Sonnet 5 $0.00102 $0.00102
Haiku 4.5 $0.00051 $0.00051

Measured yesterday against content hash 2758b048c6f4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

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.

.cursor/rules/workflow.mdc · 54 lines

What it actually says

Workflow

Before starting any non-trivial task (3+ steps or touches architecture)

  1. Stop. Write a PRD to prds/<feature-name>.md before touching code.
    • PRD files end in .md, not .prd
    • Include: problem, proposed solution, files to change, edge cases
    • For bug fixes: problem, root cause, fix, verification steps
  2. Add tasks to TASK.md under ## To Do as checkable items
  3. If anything is unclear, ask before implementing

During implementation

  • Work task by task. Mark [x] as you go in TASK.md
  • Never mark a task complete without verifying it works
  • Use subagents for research, exploration, and parallel analysis
  • One task per subagent for focused execution
  • When given a bug: just fix it. Find the root cause in logs/errors/types. No handholding needed.
  • If something goes sideways: STOP and re-plan. Don't keep pushing.
  • Before shipping: ask yourself "Would a staff engineer approve this?"

After every feature or fix

Always update both project docs. No exceptions:

  1. TASK.md - Move completed items to ## Completed with bullet details and date
  2. changelog.md - Add keepachangelog.com format entry. Run git log --date=short -n 10 for real dates. No placeholders, no future months.
  3. files.md - Update if new files were added or descriptions changed

Self-improvement

  • After any user correction: note the pattern in prds/lessons.md to prevent repeat
  • If a fix feels hacky: pause and implement the elegant solution instead
  • Simple obvious fixes: skip the reflection, just fix it
  • Review prds/lessons.md at the start of sessions when relevant

PRD rules

  • All PRDs live in prds/ folder
  • File name: prds/<feature-or-problem-slug>.md
  • Extension is always .md not .prd

Subagent strategy

  • Offload research and exploration to subagents to keep the main context clean
  • Throw more compute at complex problems via subagents
  • Use one subagent per focused task, not one subagent for everything
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 · 54 lines · 512 tokens per session scan A 2758b048c6f4

Subscribe to this mod's changes

workflow is a cursor rule published in the GitHub repository waynesutton/markdown-site (631 stars, last pushed 3mo ago), licensed MIT. It adds 512 tokens to every session, about $0.0026 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.