validation-first-driven

validation-first-driven is a skill for Claude Code, Codex from OutlineDriven/odin-claude-plugin. It costs 78 tokens per session (1,267 once invoked), scanned A, original, Apache-2.0.

A development method for systems with explicit states, transitions, or time-dependent rules. It can express high-risk designs as formal models checked by tools such as TLA+ or Alloy.

In plain words
What is it for?
Use it for protocols, workflows, concurrent systems, or other features whose behavior changes through a lifecycle.
Why use it?
It makes hidden state changes and rules explicit, helping reveal invalid transitions and broken guarantees before they reach production.

Skill for Claude CodeCodex

Written for Claude Code and Codex: shipped in a Claude Code plugin, but also agents/openai.yaml present.

Part of the odin-code-advanced plugin — 50 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/outlinedriven/odin-claude-plugin/validation-first-driven
Any agent
npx skills add OutlineDriven/odin-claude-plugin --skill validation-first-driven
Clone the repo
git clone --depth 1 https://github.com/OutlineDriven/odin-claude-plugin

Made for: Claude Code, Codex.

Or install odin-code-advanced, the plugin that ships this one along with the rest of its 50 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 validation-first-driven

README.md
[![agentmods](https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/validation-first-driven.svg)](https://agentmods.dev/skills/outlinedriven/odin-claude-plugin/validation-first-driven)
Your own site
<a href="https://agentmods.dev/skills/outlinedriven/odin-claude-plugin/validation-first-driven"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/validation-first-driven.svg" alt="Measured on agentmods" height="20"></a>
Per session 78 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,267 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 $0.00078 $0.01267
Opus 5 $0.00039 $0.00633
Sonnet 5 $0.00016 $0.00253
Haiku 4.5 $0.00008 $0.00127

Measured 2d ago against content hash 06c706a44f22, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

validation-first-driven 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.

plugins/odin-code-advanced/skills/validation-first-driven/SKILL.md · 64 lines

How it starts

The opening of the file, as written. The whole thing — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Validation-first development

Contract

Field Bound contract
Trigger The work builds protocols, workflows, concurrent systems, or lifecycle-heavy state, or needs temporal properties or a state machine.
Authority Reversible local: no file, VCS, credential, paid, published, deployed, or remote mutation outside the stated target.
Side effect Writes the model and spec plus implementation assertions and tests; creates a TLA+ or Alloy spec for high-risk designs.
Done The state space is explicit, transitions are total, invariants are checked, and each temporal property maps to an assertion, test, or model-checker run.

Inputs

  • Requirements or design document. Required.
  • Current codebase. Optional.
  • Support files: approaches, examples, formal-tools, event-sourcing reference notes. Optional; the skill is self-contained without them.

Procedure

  1. Identify scope. Confirm the work involves explicit states, transitions, temporal properties, or lifecycle-heavy state. If it is a stateless endpoint, pure data transformation, simple CRUD without lifecycle, configuration parse, or stateless batch process, stop and return the work unaltered. Done when: the work is confirmed as state-heavy or returned unaltered.
  2. Capture states and transitions. Extract all named states, state variables, actions, guards, and side effects from the requirements. Identify every temporal property: "always eventually", "never", "until", "leads-to". Done when: all states, transitions, and temporal properties are extracted.
  3. Choose mechanism level.
    Level Mechanism Use when
    Typestate Generic type params, phantom data Protocol APIs, builder patterns, Rust FFI
    Statecharts Nested states, parallel regions Game state, multi-modal UI, XState
    Flat FSM Enum + match/switch Order lifecycle, connection management
    Actor model Independent entities, message passing Distributed systems, Erlang/Elixir
    Default: use the strongest mechanism the language supports. Done when: one mechanism level is chosen with rationale.
  4. Write state machine specification. Use this template:
    STATE MACHINE: <Name>
      STATES: S1 | S2 | S3
      VARIABLES: var1: type, var2: type
      INIT: var1 = val, state = S1
      ACTION name(args): PRE: guard -> POST: new_state, effects
      INVARIANT: condition_that_always_holds
    
    Done when: the specification is written with all states, variables, actions, and invariants.
  5. Define validation level. Rank each invariant by enforcement strength: Type system (strongest) > State machine > Contract > Runtime check (weakest). Done when: every invariant has a validation level.
  6. Encode compile-time properties in types. Use typestate, sealed classes, or discriminated unions to make invalid states unrepresentable. Encode every invariant the type system can express. Done when: every type-expressible invariant is encoded.
  7. Layer state machine modeling. For properties types cannot express, define a state machine with explicit states and transitions. For high-risk designs, write a TLA+ or Alloy spec and run the model checker. High-risk threshold: the design involves distributed consensus, lock-free concurrency, a protocol with liveness or safety guarantees, or multi-party state coordination. Designs below this threshold use the state machine spec and tests without a model checker. Done when: every non-type-expressible property has a state machine or model-checker spec.
  8. Check every transition. Verify invariants hold after each action. Confirm exhaustive matching on all states. Block on any invariant violation. Done when: every transition is checked and invariants hold.
  9. Write assertions and tests. Map each temporal property to an assertion, a test, or a model-checker run. Every transition must have a test. Every invariant must have a verification point. Done when: every temporal property, transition, and invariant has a verification point.
  10. Implement. Mirror the specification exactly: one state type, one transition function, one invariant check per concern. Keep state and behavior together. Done when: the implementation mirrors the specification.

Read the full file on GitHub · 64 lines

Files

What ships with it

5 files 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. 2d ago First seen · 64 lines · 78 tokens per session scan A 06c706a44f22

Subscribe to this mod's changes

validation-first-driven is a skill published in the GitHub repository OutlineDriven/odin-claude-plugin (35 stars, last pushed today), licensed Apache-2.0. It adds 78 tokens to every session and 1,267 once invoked, about $0.0004 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 2026-09-04.

Related

Other skills, from other repositories

java-clean-arch

Reviews or implements Clean Architecture / Hexagonal Architecture (Ports & Adapters) and DDD tactical patterns for Java projects. Use when user asks to "apply clean architecture", "implement hexagonal architecture", "add ports and adapters", "apply DDD", "refactor to clean arch", "review architecture", or "add value…

ducpm2303/claude-java-plugins · 68 tokens

java-logging

Reviews Java logging for SLF4J best practices, MDC context, structured logging, and PII safety. Use when user asks to "review logging", "check my logs", "logging review", "is my logging correct", "MDC setup", or "check for PII in logs".

ducpm2303/claude-java-plugins · 60 tokens

java-security

Reviews or implements Spring Security configuration — JWT authentication, OAuth2, method-level security, CORS, and CSRF. Use when user asks to "add authentication", "secure this API", "implement JWT", "configure Spring Security", "add OAuth2 login", "protect endpoints", or "review security config".

ducpm2303/claude-java-plugins · 63 tokens

java-adr

Creates, lists, and manages Architecture Decision Records for Java projects. Use when user asks to "create an ADR", "document this decision", "write an architecture decision", "add ADR", "list decisions", "show ADRs", or "record this architectural choice".

ducpm2303/claude-java-plugins · 54 tokens

java-health

Runs a holistic code health check scoring Security, Tests, Performance and Quality with A-F grades. Use when user asks to "check health", "score this project", "health check", "how good is this code", "overall assessment", or "code quality score".

ducpm2303/claude-java-plugins · 54 tokens

java-test

Generates JUnit 5 and Mockito unit tests or Testcontainers integration tests, auto-detecting project setup. Use when user asks to "write tests", "generate tests", "add unit tests", "create test class", "test this service", or "write integration tests".

ducpm2303/claude-java-plugins · 57 tokens