spec-driven-development

A way to plan a substantial software change by writing a short specification before coding. The specification covers the goal, approach, important decisions, edge cases, failure cases, and what is out of scope.

In plain words
What is it for?
Use it for features or changes that span multiple files, involve important design choices, or would be expensive to get wrong.
Why use it?
It catches wrong assumptions while they are still easy to change, before they become code spread across many files. Agreement on the specification also limits scope growth during implementation.

Skill for Claude CodeCodex

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/buildmoonshot/skillpacks/spec-driven-development
Any agent
npx skills add buildmoonshot/skillpacks --skill spec-driven-development
Clone the repo
git clone --depth 1 https://github.com/buildmoonshot/skillpacks

Made for: Claude Code, Codex.

Per session 66 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 461 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 $0.00066 $0.00461
Opus 5 $0.00033 $0.00230
Sonnet 5 $0.00013 $0.00092
Haiku 4.5 $0.00007 $0.00046

Measured yesterday against content hash 386762076e99, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

spec-driven-development 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.

skills/engineering/expert/spec-driven-development/SKILL.md · 31 lines

What it actually says

Spec-Driven Development

For substantial work, write a short spec and agree on it before writing code. The design decisions are cheapest to change as text, before they're encoded across a dozen files.

Write the spec first

Keep it short — a focused page, not a document. Cover:

  1. Goal — what this needs to accomplish, in one or two sentences. What does "done" look like?
  2. Approach — the high-level shape: the main pieces, where they live, how data flows. Enough that someone could predict the file structure.
  3. Key decisions — the choices that matter, each with a one-line rationale: data model, the library or pattern picked, API shape, where state lives. Note the alternatives you rejected and why.
  4. Edge cases & failure modes — the inputs and conditions that will break a naive version (empty, concurrent, error, large, malicious). How each is handled.
  5. Out of scope — what this deliberately does not do, so scope doesn't creep mid-build.

Get alignment, then build

Present the spec and confirm direction before implementing. A wrong assumption caught in the spec costs one sentence; caught after implementation it costs a rewrite. Once agreed, the spec becomes the checklist you build and verify against.

Match the effort to the work

This is for substantial, multi-file, or hard-to-reverse work. A one-file change doesn't need a spec — that's what lightweight planning is for. Don't turn every task into a design doc; reserve this for when the design genuinely carries risk.

Why this matters

The most expensive mistakes are architectural — chosen early, discovered late, and woven through everything by the time they surface. A short spec surfaces those decisions while they're still one paragraph to change, and turns implementation into execution against a plan instead of design-by-accident.

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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 · 31 lines · 66 tokens per session scan A 386762076e99

Subscribe to this mod's changes

spec-driven-development is a skill published in the GitHub repository buildmoonshot/skillpacks (2 stars, last pushed 2mo ago), licensed MIT. It adds 66 tokens to every session and 461 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.

Related

Other skills, from other repositories

bridge

Use when the user wants hyperflow's behavioral rules to apply outside the terminal CLI — in Claude Code Desktop, claude.ai web, or IDE extensions that don't load CLI plugins. Writes a managed doctrine block into the project's CLAUDE.md so autonomy + intent-routing + commit cadence + role separation + file-first rules…

jeremylongshore/tons-of-skills-marketplace · 126 tokens

fable-method

A step-by-step problem-solving loop (classify the ask, define done, gather evidence, decide, act surgically, verify by observation, report outcome-first). Use when the user says "/fable-method", "use the fable method", or "approach this like Fable", or proactively when starting any multi-step task that no…

Sahir619/fable-method · 105 tokens

fable-domain

Discuss a domain with the user, research it from real sources, then generate a trusted skill bundle for it - a step-by-step workflow with a flowchart, a domain adapter, a trap fixture, and a smoke eval. Use when the user says "/fable-domain ", "make a skill for ", "add a domain to the fable method", or "give a lesser…

Sahir619/fable-method · 115 tokens

caliber-testing

Writes Vitest tests following project patterns: tests/ directories, vi.mock() for module mocking with vi.hoisted() for test-time factories, global LLM mock from src/test/setup.ts, environment variable save/restore in beforeEach/afterEach, vi.clearAllMocks() lifecycle, and test file organization. Use when user says…

caliber-ai-org/ai-setup · 113 tokens

writers-pattern

Add a new platform writer module in src/writers/ that generates and writes agent config files for a supported platform. Each writer exports a function that accepts a config interface, creates directories (rules/, skills/, mcp configs), writes files with proper formatting and frontmatter, and returns string[] of…

caliber-ai-org/ai-setup · 111 tokens

llm-provider

Adds a new LLM provider implementing LLMProvider interface with call() and stream() methods. Integrates with provider factory in src/llm/index.ts, config detection in src/llm/config.ts, and error handling via tracking and recovery. Use when adding a new model backend, integrating a third-party LLM API, or extending…

caliber-ai-org/ai-setup · 100 tokens