Outcome Engineering

Use when starting significant features, evaluating build-vs-buy decisions, reframing failed approaches, or when the user mentions "outcome engineering", "o16g", "outcome-first", "what's the outcome", or "frame this as an outcome". Also use proactively when a task description focuses on implementation ("build X", "add Y") without defining the desired outcome or verification criteria.

38 stars 4 forksFinance & Crypto Plugin marketplace

Install

Claude Code plugin
/plugin marketplace add sam-dumont/claude-skills

The repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add sam-dumont/claude-skills --skill outcome-engineering
Manual install
git clone --depth 1 https://github.com/sam-dumont/claude-skills.git /tmp/claude-skills
mkdir -p ~/.claude/skills
cp -r /tmp/claude-skills/plugins/outcome-engineering/skills/outcome-engineering ~/.claude/skills/outcome-engineering

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Outcome Engineering (o16g)

Based on the [o16g manifesto](https://o16g.com/) by Cory Ondrejka. Software engineering is outcome delivery, not code production.

| Task Thinking | Outcome Thinking |
|---------------|-----------------|
| "Build feature X" | "Enable users to achieve Y, verified by Z" |
| "Fix bug B" | "Restore correct behavior, prevent recurrence" |
| "Refactor module M" | "Reduce change cost from X to Y, measured by Z" |
| "Add technology T" | "Solve problem P with measurable improvement" |

---

## Phase 1: The Outcome Frame

Before starting significant work, produce an **Outcome Frame**. This takes ~3 minutes, not 30. Skip it for trivial tasks (typo fixes, obvious single-line bugs).

### 1. Define the Outcome (P01: Human Intent)

State the measurable change being delivered. Not what you'll build — what will be different when you're done.

**Anti-pattern**: "Build a REST API for user management"
**Outcome-first**: "External services can create, read, update, and delete user records over HTTP, with <200ms p95 latency"

Ask: **"What is measurably different when this is done?"**

### 2. Define Verification (P02: Verified Reality)

How will you PROVE it worked? Not "it looks right" — observable, repeatable evidence.

Verification must be concrete:
- Tests that pass/fail (unit, integration, e2e)
- Observable behavior (API returns X, UI shows Y)
- Metrics that move (latency drops, error rate decreases)
- User action that succeeds (can complete workflow Z)
Read the full skill on GitHub

Details

Skill name
outcome-engineering
Path
plugins/outcome-engineering/skills/outcome-engineering/SKILL.md
License
MIT
Repo updated
3 November 2026

FAQ

What does the Outcome Engineering skill do?

Use when starting significant features, evaluating build-vs-buy decisions, reframing failed approaches, or when the user mentions "outcome engineering", "o16g", "outcome-first", "what's the outcome", or "frame this as an outcome". Also use proactively when a task description focuses on implementation ("build X", "add Y") without defining the desired outcome or verification criteria.

How do I install the Outcome Engineering skill?

Claude Code plugin: /plugin marketplace add sam-dumont/claude-skills. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Outcome Engineering popular?

It lives in sam-dumont/claude-skills, which has 38 GitHub stars and 4 forks.

Related skills

Advanced Code Quality

38

sam-dumont/claude-skills

Use when going beyond basic linting to enforce advanced Python code quality gates. Triggers on: cognitive complexity, code duplication detection, architectural enforcement, import boundaries, dependency hygiene, test quality beyond coverage, mutation testing, docstring coverage, type coverage, AI-generated code detection, AI slop detection, complexipy, jscpd, import-linter, deptry, diff-cover, mutmut, interrogate, sloppylint, refurb, pylint design checks, wily, radon maintainability, pytestarch, griffe, "bulletproof Python", "tighten quality gates", "advanced quality checks", "beyond ruff and

Software Development

Code Quality

38

sam-dumont/claude-skills

Automatically triggered when setting up Python code quality tooling, configuring linters, adding type checking, improving code standards, running code quality checks, setting up pre-commit hooks, or auditing Python code for quality issues. Applies when discussing: ruff setup, mypy configuration, code formatting, linting rules, cyclomatic complexity, dead code detection, file length limits, pre-commit hooks, Makefile targets for quality, uv-based workflows, pyproject.toml configuration, or "check my code quality". Also triggers on: "set up linting", "add type checking", "configure ruff", "forma

Software Development

Code Security

38

sam-dumont/claude-skills

Automatically triggered when running security scans, auditing code for vulnerabilities, checking dependencies for CVEs, reviewing code for security issues, or hardening Python applications. Applies when discussing: security audit, vulnerability scan, bandit, pip-audit, dependency vulnerabilities, OWASP, injection, path traversal, secrets detection, insecure deserialization, auth bypass, or "is this code secure?". Also triggers on: "run security checks", "find vulnerabilities", "security review", "check for CVEs", "audit dependencies", "security lint", "pen test this code", "harden this applica

Security

Rubber Duck

62

ElliotJLT/Claude-Skill-Potions

When a user is stuck, frustrated, or describing a problem vaguely, do NOT immediately suggest solutions. First, force structured problem articulation through targeted questions: What did you expect? What happened instead? What have you tried? Only after the problem is clearly defined, propose solutions.

Finance & Crypto

Tax Fraud

44

Jakeschincariol/unhinged-claude-skills

Hunts cost everywhere — trims tokens, kills redundant calls, and reports the cheapest path to the same result; invoke via /tax-fraud when you want aggressive (but technically legal) cost optimization of an AI workflow, pipeline, or process.

Finance & Crypto13

Diagnose Errors

48

amplitude/mcp-marketplace

Investigates errors across network failures, JavaScript errors, and error clicks to identify what's broken, where, and why. Use when the user says "what's broken", "errors are up", "why are users seeing errors", "JS errors", "network failures", "5xx spike", "something is broken", or wants to triage product reliability issues.

Finance & Crypto12

Browse the most popular Claude skills, all finance & crypto skills, or write your own with the skills generator.