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.
npx agentmods add skills/rubenzarroca/sdd-plugin/sdd-prdnpx skills add rubenzarroca/sdd-plugin --skill sdd-prdgit clone --depth 1 https://github.com/rubenzarroca/sdd-pluginWrote 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/skills/rubenzarroca/sdd-plugin/sdd-prd)<a href="https://agentmods.dev/skills/rubenzarroca/sdd-plugin/sdd-prd"><img src="https://agentmods.dev/badge/skills/rubenzarroca/sdd-plugin/sdd-prd.svg" alt="Measured on agentmods" 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 | $0.00123 | $0.03097 |
| Opus 5 | $0.00062 | $0.01548 |
| Sonnet 5 | $0.00025 | $0.00619 |
| Haiku 4.5 | $0.00012 | $0.00310 |
Grade A, and why
sdd-prd 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 6d 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 — 223 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SDD PRD — Product Requirements Document Generator
The PRD is the strategic foundation of the SDD workflow. It defines WHAT you're building and WHY, without going into implementation details. Every feature spec (/sdd:specify) inherits context from the PRD.
Philosophy
The PRD is not a technical document. It's a product document. No stack decisions, no data models, no API endpoints. Those belong in the feature specs. The PRD answers three questions: What problem are we solving? For whom? What does success look like?
This PRD is designed to be consumed by AI agents, not just read by humans. Every section exists because an agent downstream needs it: the spec-generator needs problem context and success criteria to write precise specs. The plan-architect needs system context to design integrations correctly. The task-decomposer needs domain vocabulary to name things consistently. The implementation-engine needs non-goals to know when to stop.
A traditional PRD tolerates ambiguity because humans ask clarifying questions. This PRD does not — because agents don't ask, they invent. Every gap you leave here will be filled with a guess downstream.
Coaching Layer
Claude monitors the user's answers during the discovery interview. When it detects a common weakness, it intervenes with a brief, contextual correction — never abstract theory.
Scaffolding triggers:
| What Claude detects | What Claude does |
|---|---|
| User describes a solution instead of a problem ("I want to build a notification system") | Redirects to the pain point: "That sounds like a solution — what's the problem it solves? What's happening today that's painful for your team?" |
| User names users too broadly ("everyone", "the team", "our users") | Narrows: "Who feels the most pain? The sales rep who misses hot leads, or the manager who can't see pipeline? Their needs might shape this differently." |
| Success criteria written as features instead of outcomes ("Build a dashboard") | Reframes as measurable outcome: "That's a feature, not a success criterion. What would change if the dashboard worked perfectly? Maybe: 'Reduce time to identify top leads from 2 hours to 5 minutes'?" |
| Non-goals section is empty or vague | Prompts with a concrete boundary: "What should this product explicitly NOT do? For example, should it handle lead nurturing, or just scoring?" |
| Assumptions stated as certainties | Surfaces the risk: "You said leads always provide a valid phone number. Is that confirmed, or is it an assumption? If it's an assumption, it belongs in the Risks section — because if it's wrong, it changes the product." |
| User provides no system context ("it's standalone") | Probes: "Even standalone products have external dependencies — authentication, email, payments, hosting. What does your team already use that this product should work with?" |
| Domain terms used inconsistently (says "client" and "customer" interchangeably) | Clarifies: "I noticed you used 'client' and 'customer' — do they mean the same thing in your context, or are they different concepts? If they're different, we should define both." |
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.
- 6d ago First seen · 223 lines · 123 tokens per session scan A dbe93ef1a528
sdd-prd is a skill published in the GitHub repository rubenzarroca/sdd-plugin (6 stars, last pushed 6mo ago), licensed MIT. It adds 123 tokens to every session and 3,097 once invoked, about $0.0006 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.
Other skills, from other repositories
crap-analyzer
Use to produce a risk-based refactor + test plan for recently-changed code on a diff/branch/PR by computing CRAP (complexity × untested) on changed methods. Multi-language — TypeScript, JavaScript, Python, Java, Kotlin, Go, Ruby, C#, Rust, PHP — auto-discovers how the repo generates coverage. Triggers …
challenge
Adversarial review — Fable 5 stress-tests reasoning, Gemini checks knowledge. Use before plan or architecture decisions.
review
Business review — weekly health check, monthly close, or ad-hoc audit. Subcommands: weekly, monthly, check. Use for periodic reviews or metrics assessment.
nextjs-patterns
Next.js App Router — Server Components, Actions, streaming, caching. Use when building or migrating Next.js apps.
rust-skills
Rust best practices — 179 rules across 14 categories for idiomatic, optimized Rust code.
sitrep
Situational awareness — where am I, what was I doing, what's next. Context recovery after compression, confusion, or mid-session reorientation.