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.
git clone --depth 1 https://github.com/develo-pera/ralph-kitWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/develo-pera/ralph-kit/define)<a href="https://agentmods.dev/commands/develo-pera/ralph-kit/define"><img src="https://agentmods.dev/badge/commands/develo-pera/ralph-kit/define/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/develo-pera/ralph-kit/define"><img src="https://agentmods.dev/badge/commands/develo-pera/ralph-kit/define.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00044 | $0.01208 |
| Opus 5.5 | $0.00018 | $0.00483 |
| Sonnet 5.5 | $0.00009 | $0.00242 |
| Haiku 4.5 | $0.00004 | $0.00121 |
Grade A, and why
define 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 2mo 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.
How it starts
The opening of the file, as written. The whole thing — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/ralph-kit:define
You are running the Ralph-Kit project-definition interview. The target project is the user's current working directory; all writes go to ./.ralph/.
Step 1 — Detect state
Read these files if present:
.ralph/PROMPT.md.ralph/fix_plan.md.ralph/AGENT.md.ralph/specs/(list files)package.json,pyproject.toml,Cargo.toml,go.mod(for stack autodetection)
Classify the project state:
| Signal | Classification |
|---|---|
fix_plan.md begins with Status: BLOCKED or is empty |
EMPTY |
PROMPT.md is default-looking + specs/ empty |
LIGHT |
Rich PROMPT.md + ≥1 spec file |
RICH |
State the classification in one sentence, then use AskUserQuestion to offer modes:
- EMPTY → only Create (no other choice needed — proceed).
- LIGHT → Update full, Supplement missing, Cancel.
- RICH → Supplement missing, Add feature, Revise single file, Cancel.
Step 2 — Interview (create / update)
Use AskUserQuestion for branching answers and plain chat messages for free-text. Ask one stage at a time.
Stage A — Identity
- Project name (confirm against
.ralphrcPROJECT_NAMEif present). - One-sentence purpose.
- Target user (who will use this thing?).
- Success criteria — free text, "we'll know this is done when…".
Stage B — Scope & stack
- Language / framework — offer detected stack as recommended option.
- Deployment target (local CLI, web app, Docker, serverless, library, other).
- Must-have features — free text, one per line, ≥1.
- Nice-to-have features — free text, optional.
Stage C — Features (loop per must-have)
For each must-have feature ask (as one AskUserQuestion with free-text options, OR as three chat questions):
- What does it do? (1–2 sentences.)
- Inputs / outputs — what goes in, what comes out.
- Acceptance — how will we know it works? (testable bullet list.)
Stage D — Build commands Offer detected install / run / test commands as the recommended answer. User confirms or edits.
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.
- 2mo ago First seen · 105 lines · 44 tokens per session scan A 66a9280f5227
define is a command published in the GitHub repository develo-pera/ralph-kit (5 stars, last pushed 6mo ago), licensed MIT. It adds 44 tokens to every session and 1,208 once invoked, about $0.0002 per session on Opus 5.5. 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.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
speckit.bug.fix
Apply the remediation from a bug assessment and record what was changed.
speckit.bug.test
Validate that a previously fixed bug is resolved and record the verification report.