ralph

A persistence workflow driven by a product requirements document, or PRD, which lists the work as individual stories with acceptance checks. It implements, tests, and revisits each story until all checks pass.

In plain words
What is it for?
It is for completing feature work story by story, running relevant tests, fixing failures, and performing a final review when every story is verified.
Why use it?
It helps keep multi-step development work moving when some stories fail verification and need further fixes.

Cursor rule

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/dasomel/oh-my-cursor/ralph
Clone the repo
git clone --depth 1 https://github.com/dasomel/oh-my-cursor
Per session 13 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 818 The whole file, excluding the scripts and references it only reads on demand.
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 $0.00013 $0.00818
Opus 5 $0.00006 $0.00409
Sonnet 5 $0.00003 $0.00164
Haiku 4.5 $0.00001 $0.00082

Measured 2d ago against content hash a6629b88d12e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ralph 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 2d ago.

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.

rules/workflows/ralph.mdc · 124 lines

How it starts

The opening of the file, as written. The whole thing — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Ralph Workflow

Ralph is a persistence loop that tracks work via a PRD (Product Requirements Document) and iterates until all stories pass verification.

Triggers

  • "ralph", "don't stop", "keep going", "must complete", "until done"

How It Works

1. Create PRD → .omc-cursor/prd.json
2. For each story:
   a. Implement (Executor role)
   b. Verify (Verifier role)
   c. If FAIL → fix and re-verify
3. Repeat until ALL stories pass
4. Final sign-off (Architect role)

Phase 1: PRD Creation

Create .omc-cursor/prd.json with this structure:

{
  "title": "Feature/Task Title",
  "created": "ISO-8601 timestamp",
  "stories": [
    {
      "id": "S1",
      "title": "Story title",
      "acceptance": ["criterion 1", "criterion 2"],
      "status": "pending",
      "attempts": 0
    }
  ],
  "status": "in-progress"
}

Phase 2: Story Implementation Loop

For each story with status pending or failed:

Step 1: Implement (Executor Role)

Switching to Executor role.
- Read the story's acceptance criteria
- Implement the minimum code to satisfy ALL criteria
- Run relevant tests

Step 2: Verify (Verifier Role)

Switching to Verifier role.
- Check each acceptance criterion with EVIDENCE
- Run tests → show actual output
- Run build → show actual output
- Update story status in prd.json:
  - All criteria met → status: "passed"
  - Any criterion failed → status: "failed", increment attempts

Step 3: Fix (if failed)

Switching to Debugger role (if needed) then Executor role.
- Analyze failure
- Fix the issue
- Return to Step 2

Retry Limit

  • Max 3 attempts per story
  • After 3 failures: mark as "blocked", report to user, move to next story

Phase 3: Final Sign-Off

When all stories are passed:

Switching to Architect role.
- Review overall implementation coherence
- Check for integration issues between stories
- Verify no regression in existing functionality
- Update prd.json status to "complete"

State Management

Read the full file on GitHub · 124 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. 2d ago First seen · 124 lines · 13 tokens per session scan A a6629b88d12e

Subscribe to this mod's changes

ralph is a cursor rule published in the GitHub repository dasomel/oh-my-cursor (2 stars, last pushed 5mo ago), licensed MIT. It adds 13 tokens to every session and 818 once invoked, about $0.0001 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 2026-08-31.