specwright AGENTS.md

A set of instructions for Specwright, a workflow for building applications from written specifications with review checkpoints. It covers project setup, research, design, planning, test-driven development (writing tests before implementation), building, verification, debugging, and shipping.

In plain words
What is it for?
Use it to initialize projects, research external systems, design architecture, break work into testable units, implement with TDD, investigate bugs, verify results, and prepare pull requests.
Why use it?
It gives coding agents a repeatable path from an idea to a checked implementation, while keeping decisions and requirements documented.

Instructions file for CodexOpenCode

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 instructions/obsidian-owl/specwright/agents-md
Clone the repo
git clone --depth 1 https://github.com/Obsidian-Owl/specwright

Made for: Codex, OpenCode.

Per session 1,538 This file is loaded in full into every session.
When invoked 1,538 The same file — it is already loaded in full.
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 $0.01538 $0.01538
Opus 5 $0.00769 $0.00769
Sonnet 5 $0.00308 $0.00308
Haiku 4.5 $0.00154 $0.00154

Measured 2d ago against content hash ba2e46c0d77c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

specwright AGENTS.md 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.

AGENTS.md · 102 lines

How it starts

The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Specwright

Spec-driven app development with quality gates. Ensures the user gets what they asked for.

Compatible with AI coding agents that read AGENTS.md project instructions.

Workflow

sw-init → sw-design → sw-plan → sw-build → sw-verify → sw-ship
Skill Purpose
sw-init Project setup. Creates constitution + charter. Configures gates.
sw-research Deep outward-facing research. External docs, APIs, patterns, validation. Produces referenced briefs.
sw-design Interactive solution architecture. Research, design, adversarial critic, assumption surfacing.
sw-plan Decompose design into work units with testable specs.
sw-build TDD implementation of one work unit.
sw-verify Interactive quality gates. Shows findings, validates against spec.
sw-ship Strategy-aware merge via PR.
sw-debug Investigation-first debugging. Scope → investigate → diagnose → fix/log/defer.
sw-pivot Mid-build course correction. Revises remaining tasks via architect; append-only.
sw-doctor Read-only installation health check. 13 checks, repair hints.
sw-guard Detect stack and interactively configure guardrails.
sw-status Current state and progress.
sw-adopt Explicitly adopt an existing work into the current worktree after ownership checks.
sw-learn Post-ship capture of patterns and learnings.
sw-audit Periodic codebase health check. Finds systemic tech debt.
sw-sync Git housekeeping. Fetch, prune stale branches, sync with remote.
sw-review PR comment review. Fetch all comment types, group by status, respond inline.

Anchor Documents

Three persistent tracked documents drive all decisions. By default they live under .specwright/, the tracked projectArtifactsRoot:

  • .specwright/CONSTITUTION.md -- Development practices. How the user wants code written. The AI MUST follow these.
  • .specwright/CHARTER.md -- Technology vision. What this repo is, who consumes it, architectural invariants.
  • .specwright/TESTING.md -- Testing strategy. How the project should be tested, what boundaries exist, what may be mocked. Optional — created during init if the user opts in.

Read the full file on GitHub · 102 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. 2d ago First seen · 102 lines · 1,538 tokens per session scan A ba2e46c0d77c

Subscribe to this mod's changes

specwright AGENTS.md is an instructions file published in the GitHub repository Obsidian-Owl/specwright (9 stars, last pushed 4mo ago), licensed MIT. It adds 1,538 tokens to every session, about $0.0077 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.