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.
npx agentmods add instructions/onehillai/asdd/agents-mdgit clone --depth 1 https://github.com/OneHillAI/ASDDWhat 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 | $0.01036 | $0.01036 |
| Opus 5 | $0.00518 | $0.00518 |
| Sonnet 5 | $0.00207 | $0.00207 |
| Haiku 4.5 | $0.00104 | $0.00104 |
Grade A, and why
ASDD 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.
How it starts
The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
This is ASDD constitution template. Copy it to
AGENTS.mdin the root of your repo and adapt the sections marked (adapt). It is the tool-agnostic standing-instructions file every agent reads - Goose, Claude, Copilot, Cursor - and tool-specific files likeCLAUDE.mddefer to it. The filled-in reference copy is shipped in this repository. Keep the sections marked (fixed) as they are - they are what makes a change conformant.
Project and purpose (adapt)
One paragraph: what this project is and the outcome it exists to produce. Link your authoritative design doc so an agent starts from the real architecture, not a guess.
Build, run, test (adapt)
The commands an agent needs: install, run, the test command, the lint/format/type checks. CI runs the same and blocks merge. Name the local model or services if the tests need them.
Working protocol (house style)
- Explorative but factual - verify external claims against real sources or code; separate verified from assumed.
- Cross-check after every commit - does it run, drift from the design, or leave leftovers? Fix in a follow-up.
- No AI slop - lean, human-idiomatic code; no narration, padding, or speculative abstraction.
Follow the repo's conventions (Conventional Commits, a test in the style of the existing suite, a changelog note for user-facing changes; plain hyphens if the project lints for them).
Lanes (adapt the set; keep "exactly one per PR")
Tag every PR with exactly one lane; the intake gate enforces it, reading the set from .asdd.yml
(lanes:). The reference set, swap in your own:
feature·fix·docschore(trivial change; skips the spec requirement)
Give any lane a charter of invariants a change must not weaken if you want one. (Link your charters.)
The PR contract (fixed - four items)
Every PR, from a human or an agent, needs the same four things:
- Disclosure - tick human or AI-agent in the PR template; an agent's commits carry
Agent: <name> (automated, instructed-by: <human>). - Exactly one lane label.
- DCO sign-off -
git commit -son every commit. - A test for new behaviour.
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.
- 2d ago First seen · 88 lines · 1,036 tokens per session scan A b9b6f0934896
ASDD AGENTS.md is an instructions file published in the GitHub repository OneHillAI/ASDD (5 stars, last pushed 21d ago), licensed Apache-2.0. It adds 1,036 tokens to every session, about $0.0052 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.
Other instructions, from other repositories
agentic-playwright selectors.instructions.md
Instructions for idavidov13/agentic-playwright, covering selector strategy, critical, instructions, phase 1: open and authenticate and phase 2: explore like a user.
toh-framework CLAUDE.md
Instructions for wasintoh/toh-framework, covering claude.md — toh framework (repo development guide), what this is, everyday commands, verification protocol and single source, transformed per ide.
hatch3r CLAUDE.md
Instructions for hatch3r/hatch3r, covering hatch3r — development instructions, architecture, development commands, two-axis pillar framework (2.0.0) and orchestrator self-discipline (bypass protection).
AgentGo AGENTS.md
Instructions for yeasy/AgentGo, covering agents.md, purpose, startup instructions, trust & safety and failure modes.
elephant-agent copilot-instructions.md
Copilot instructions for agentic-in/elephant-agent, covering github copilot instructions, source of truth order, review priorities, harness-specific checks and validation expectations.
harmont-cli CLAUDE.md
Instructions for harmont-dev/harmont-cli, covering testing, documentation, dsl and keep the sdk, hm init templates, and docs in sync.