sdd-plan

sdd-plan is a skill for Claude Code from rubenzarroca/sdd-plugin. It costs 67 tokens per session (2,901 once invoked), scanned A, original, MIT.

A design aid for choosing how to build a specified software feature. It produces a technical plan and an architecture decision record, which documents an important design choice and why it was made.

In plain words
What is it for?
Use it to design the architecture, components, data flow, and technical decisions for a feature after its requirements have been clarified.
Why use it?
It turns an agreed feature description into a build approach that developers can evaluate before implementation begins.

Skill for Claude Code

Written for Claude Code: argument-hint in frontmatter.

Part of the sdd plugin — 12 skills shipped together

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 skills/rubenzarroca/sdd-plugin/sdd-plan
Any agent
npx skills add rubenzarroca/sdd-plugin --skill sdd-plan
Clone the repo
git clone --depth 1 https://github.com/rubenzarroca/sdd-plugin

Made for: Claude Code.

Or install sdd, the plugin that ships this one along with the rest of its 12 skills.

Wrote 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.

agentmods badge for sdd-plan

README.md
[![agentmods](https://agentmods.dev/badge/skills/rubenzarroca/sdd-plugin/sdd-plan.svg)](https://agentmods.dev/skills/rubenzarroca/sdd-plugin/sdd-plan)
Your own site
<a href="https://agentmods.dev/skills/rubenzarroca/sdd-plugin/sdd-plan"><img src="https://agentmods.dev/badge/skills/rubenzarroca/sdd-plugin/sdd-plan.svg" alt="Measured on agentmods" height="20"></a>
Per session 67 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,901 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 5.1 $0.00067 $0.02901
Opus 5 $0.00034 $0.01451
Sonnet 5 $0.00013 $0.00580
Haiku 4.5 $0.00007 $0.00290

Measured 6d ago against content hash 247196b986d7, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

sdd-plan 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.

skills/sdd-plan/SKILL.md · 255 lines

How it starts

The opening of the file, as written. The whole thing — 255 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/sdd:plan — Design the technical approach

You are designing the technical approach for a feature that has been fully specified and clarified. Follow these steps exactly, in order. Do NOT skip steps. Do NOT read source code — only spec.md, constitution.md, and state.json. Source code analysis belongs to the tasks phase.

Coaching Layer

In sdd-plan, the user shifts from input provider to output consumer — they must evaluate a technical document they didn't write. Claude must bridge this gap.

Rules for presenting technical content:

  1. Translate to business impact — calibrated to experience. Read .sdd/state.json field completed_features. On the first feature (0), translate every technical concept to business impact: not "SSR has better SEO" but "Server-side rendering means Google can read your pages." On subsequent features, translate only novel or complex concepts — the user has built vocabulary through previous features. Never re-explain concepts the user has already engaged with.
  2. Explain trade-offs in functional terms. Not "higher coupling" but "these two pieces depend on each other, so changing one means changing both."
  3. Give the user evaluation criteria — calibrated. Read .sdd/state.json field completed_features.
    • 0: Full evaluation criteria: "When reviewing this plan, focus on three things: (1) Does the data flow match how you described the feature? (2) Are there risks that feel too high for this stage? (3) Is there anything you don't understand? It's better to ask now than to discover a misunderstanding during implementation." Then add: "The more you engage here — catching risks, questioning decisions, suggesting changes — the more I calibrate to your level and move faster in future features."
    • 1: Brief reminder: "Same three things to check — data flow, risks, and anything unclear."
    • 2+: Skip the criteria. Just ask: "Does this match what you had in mind?"
  4. Explain what an ADR is on the user's first feature. Check .sdd/state.json field milestones.adr_explained. If false, explain: "An ADR (Architecture Decision Record) is a short document that records why we chose this approach. It's useful when someone asks 'why was it built this way?' six months from now." Then set milestones.adr_explained to true in state.json. If already true, skip the explanation.

Read the full file on GitHub · 255 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. 6d ago First seen · 255 lines · 67 tokens per session scan A 247196b986d7

Subscribe to this mod's changes

sdd-plan is a skill published in the GitHub repository rubenzarroca/sdd-plugin (6 stars, last pushed 6mo ago), licensed MIT. It adds 67 tokens to every session and 2,901 once invoked, about $0.0003 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.