What is dotdog AGENTS.md?

A project guide for the dotdog command-line tool, explaining how agents should query its machine-readable specification graph and verify code changes.

Use it when exploring dotdog entities and relationships, changing dotdog source, updating specifications, or running targeted and full tests.

It prevents agents from reading or editing the human-format specification directly and keeps changes tied to the project’s source of truth.

dotdog AGENTS.md — Instructions file by specdog

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/specdog/dotdog/agents-md

Clone the repo

git clone --depth 1 https://github.com/specdog/dotdog

Made for: Codex, OpenCode.

Per session 509 This file is loaded in full into every session.

When invoked 509 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.

Model	Per session	Once invoked
Fable 5	$0.00509	$0.00509
Opus 5	$0.00254	$0.00254
Sonnet 5	$0.00102	$0.00102
Haiku 4.5	$0.00051	$0.00051

Measured yesterday against content hash 61322e71c825, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

dotdog 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 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.

AGENTS.md · 50 lines

What it actually says

AGENTS.md — dotdog

This repo is the dotdog CLI tool. For spec graphs, use npx dotdog serve (MCP) or dotdog compile.

Quick Start (read this first)

NEVER read .dog files directly. Query the .dag via MCP. The .dog is human format. The .dag is agent format.

.md: read for context (README, docs, blog) but NOT for entity/relationship/project structure
.dog: NEVER read. Human writes it. Agent never touches it.
.dag: ALWAYS query via MCP. This is your source of truth for project structure.
If .dag doesn't have it: report "unverified" — never fall back to .dog

MCP: npx dotdog serve (7 tools: getEntity, traverse, search, schema, summary, listProjects, infraVerify)

Workflow (lean):

1 issue → 1 PR → 1 merge
spec/.dag first, code second
no unrelated edits
verify with targeted test, then full bun test
update docs/spec in the same PR if behavior changes
never merge without explicit user ask

Token efficiency:

keep truth in .dag/spec, not chat
preserve IDs, file paths, and test results verbatim
compress chatter/logs, not constraints or state
prefer short status: changed / verified / next

Public site/docs: keep README, docs, and specdog.github.io in sync when adding commands, docs, blog posts, or discovery pages.

Commands (when modifying dotdog source)

dotdog validate     → check completeness (run before commits)
dotdog doctor       → baseline health check (run before commits)
dotdog analyze      → gap detection + contradiction checking
dotdog compile      → rebuild .dag from .dog files
dotdog index        → build search index
dotdog search "q"   → semantic search across specs
dotdog staleness    → detect spec/reality drift

Key Rules

Use conventional commits (feat:, fix:, docs:, chore:)
Never commit dist/
Every code change touching packages/dotdog/src/ or package.json must bump version
Dogfood: test against npx dotdog@latest validate before shipping

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.

yesterday First seen · 50 lines · 509 tokens per session scan A 61322e71c825

Subscribe to this mod's changes

dotdog AGENTS.md is an instructions file published in the GitHub repository specdog/dotdog (4 stars, last pushed 1mo ago), licensed MIT. It adds 509 tokens to every session, about $0.0025 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.

Planning Testing ai cli dogfood dotdog spec specification

dotdog AGENTS.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

What it actually says

AGENTS.md — dotdog

Quick Start (read this first)

Commands (when modifying dotdog source)

Key Rules

What this file has done since we first saw it

Other instructions, from other repositories

agents

claude

spec-kit AGENTS.md

comet CLAUDE.md

agent-spec AGENTS.md

modelcontextprotocol CLAUDE.md