Evidence Driven Engineering — AI skill for Claude Code
A lightweight evidence-first engineering judgment framework (Agent Skill) for AI coding agents — debugging, code review, legacy maintenance, and incident analysis without confusing plausibility for tr.
How to install Evidence Driven Engineering
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open doulos76/evidence-driven-engineering and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Evidence Driven Engineering does
A lightweight evidence-first engineering judgment framework (Agent Skill) for AI coding agents — debugging, code review, legacy maintenance, and incident analysis without confusing plausibility for truth.
Alternatives in AI
- Agentic SEO Skill — An LLM-first SEO analysis skill for Antigravity, Codex, Claude with 16 specialized sub-skills, 10 specialist a 419 ★
- Openase — Ticket-Driven Automated Software Engineering 265 ★
- Ring — 89 skills and 38 specialized agents that enforce proven engineering practices for AI-assisted development 176 ★
README
Evidence Driven Engineering (EDE)
**Teach your coding agent not to confuse plausibility with truth.**
Evidence Driven Engineering is a lightweight engineering judgment framework for AI coding agents.
It helps an agent:
- distinguish facts from assumptions
- avoid anchoring on the first plausible root cause
- search for evidence that disproves its own hypothesis
- understand legacy context before changing code
- prefer the smallest supported change
- verify behavior before claiming success
- communicate remaining uncertainty
Core Principles
**Evidence before conclusions.** **Falsification before confidence.** **Context before modification.** **Verification before claims.**
And:
**Be strict with evidence, humble with interpretation, and skeptical of your own conclusions.**
What EDE Is Not
EDE is not a replacement for TDD, code review, static analysis, or your development process.
It is a reasoning layer that complements them.
Repository Status
v0.3.0 is implemented and published. `SKILL.md` is complete, and all 13 behavioral scenarios in `tests/scenarios.md` have been run and recorded — see [Behavioral Testing](#behavioral-testing) below.
See [PRD.md](./PRD.md) for the full product requirements.
Repository Structure
evidence-driven-engineering/
├── SKILL.md
├── README.md
├── PRD.md
├── CONTRIBUTING.md
├── LICENSE
├── CHANGELOG.md
├── TODO.md
├── .github/
│ ├── ISSUE_TEMPLATE/
│ │ ├── bug_report.md
│ │ └── feature_request.md
│ ├── PULL_REQUEST_TEMPLATE.md
│ ├── markdown-link-check.json
│ └── workflows/
│ └── ci.yml
├── docs/
│ └── CODEX_HANDOFF.md
├── scripts/
│ ├── check_test_record.py
│ ├── package_skill.py
│ └── validate_skill.py
├── references/
│ ├── examples.md
│ └── anti-patterns.md
└── tests/
├── scenarios.md
├── rubric.md
└── results/
├── README.md
└── 2026-08-21-*.md
Example Prompts
Use evidence-driven-engineering to investigat
Related Skills
Common Sense Skill
Portable agent skill enforcing goal-first judgment, existing-capability discovery, free-first cost checks, evi
Autonomous Debugger
Evidence-driven debugging skill for Claude Code and AI coding agents. Investigates codebases, reproduces failu
Universal Debug Mode Skill
Evidence-driven debugging skill for AI agents: generate competing hypotheses, instrument with isolated runtime
Secondmate Reason
Ask a read-only reasoning model (via secondmate) a hard analysis question — root-cause, triage, plan review, p
Agentic Engineering Protocol
Verification-driven operating protocol for AI coding agents - Explore/Plan/Implement/Verify/Deliver with adver
Supergraph
Engineering workflow system for AI coding agents — mandatory planning, TDD, verification, review, and intellig
Related Agents
Swe Dev Loop Systems Architect
Designs and audits coding-agent loops, debugging and migration workflows, CI/release loops, autonomous-science
Hephaestus
Autonomous deep implementation worker. Full tool access for complex coding tasks. Explores codebase patterns,
Support Liaison
Translates raw customer input (angry email, confusing Slack thread, ticket chain, Loom transcript, support cal