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/junghan0611/entwurf/agents-mdgit clone --depth 1 https://github.com/junghan0611/entwurfWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/junghan0611/entwurf/agents-md)<a href="https://agentmods.dev/instructions/junghan0611/entwurf/agents-md"><img src="https://agentmods.dev/badge/instructions/junghan0611/entwurf/agents-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.07591 | $0.07591 |
| Opus 5 | $0.03795 | $0.03795 |
| Sonnet 5 | $0.01518 | $0.01518 |
| Haiku 4.5 | $0.00759 | $0.00759 |
Grade A, and why
entwurf 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 today.
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 — 225 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — Maintainer Guidelines for entwurf
For agents that own this repo: invariant principles and reproducible verification, not release-story archaeology.
Direction. This repo is the entwurf capability package: a v2 garden-citizen dispatch substrate, native-harness bridges, an ACP plugin, and the pi adapter that hosts that plugin today.
entwurfis the subject; pi is one harness adapter. ACP is a plugin, not the boundary (#38). V1 verbs are gone. V2 addresses existing citizens; fresh sibling creation is the separateentwurf_fresh_callcapability. Current work and ordering live in NEXT.md.
North Star — One Forged Screwdriver
entwurf는 스위스 아미 나이프나 두 번째 하네스가 아니다. 이것은 담금질된 드라이버 한 자루다: 작고, 명시적이고, 맡은 접점에서만 강해야 한다.
- entwurf가 주어이고 pi는 한 adapter다. pi는 가장 깊게 붙은 하네스지만 다른 하네스보다 높은 정체성 계층이 아니다.
- 다른 하네스의 세션은 형제다. Claude Code, Codex, Antigravity, pi는 각자의 transcript/auth/runtime을 소유한다. 증명된 lifecycle과 transport가 있을 때 같은 garden address space의 citizen이 된다.
- 능력을 surface 이름으로 재단하지 않는다. 도구 이름이나 transport 비대칭은 capability 차이이지 존재의 등급이 아니다.
- substrate는 결정적 dispatch만 맡는다. record에서 identity를 읽고, rail별 liveness와 caller intent로 transport를 고른다. 숨은 hydration, ambient MCP scanning, 근거 없는 tool claim을 만들지 않는다.
- 좁은 tool surface는 규율이다. entwurf가 backend를 몰 때 sub-agent/todo 없이 한 자루 드라이버로 움직인다. 두 번째 orchestrator로 자라지 않게 한다.
- entwurf는 부속품이 아니라 분신을 연다. resumed/spawned session은 runtime-isolated peer이지 disposable worker가 아니다.
- 증거가 말을 훈육한다. README, source, gates, VERIFY, BASELINE이 받치지 않는 강한 주장은 멈춘다.
판단할 때 묻는다:
- tool 이름을 보는가, capability를 보는가?
- backend 비대칭을 정직하게 기록하는가, 형제성을 포기하는 핑계로 쓰는가?
- 두 번째 하네스를 만드는가, 드라이버 한 자루를 단단하게 만드는가?
- 이미 주어진 방향을 되묻는가, 실행 가능한 다음 증거를 만드는가?
Architecture
- entwurf-core (v2) owns garden-id addressing, peer facts, liveness interfaces, dispatch decisions, rail choice, and delivery evidence.
- Record authority owns citizen identity. Every addressable citizen uses the same V3
MetaIdentityschema.backendselects capability/rail behavior; it does not create an identity hierarchy. - pi adapter attaches a pi session to a record at
session_start, hosts the record-keyed control socket, and exposes the native pi tool surface. - Native bridges register already-running native sessions without taking over their transcript or auth: Claude Code and Copilot CLI are mailbox/self-fetch (a Claude hook arms the watch; a forked Copilot extension holds it); Antigravity is probe-backed native-push. Codex has archived probe evidence, but its managed native lane was declined because pi already supplies the official GPT provider path; do not duplicate it as a native citizen or ACP backend.
- ACP plugin registers provider
entwurfinside a pi host session and drives a backend under an isolated overlay. The host pi session is already a record-backed socket citizen; the plugin does not mint another citizen/socket/peer layer. - mux is launch-only and deliberately small. Shipped: the tmux placement leaf (
mux-placement.ts— inspect/append/close inside the caller's own session), the visible runtime launch composition (mux-launch.ts), the fresh-call composition (mux-fresh-call.ts), and the cwd-aware resume placement composition (mux-resume-call.ts). Fresh call is exposed asentwurf_fresh_call; visible same-id resume isentwurf_resume_call, composed at the two public surfaces by injectingmux-resume-callinto the v2-sideentwurf-v2-visible-resume.ts. The exact import graph is enumerated in docs/mux-launch-rail.md §11.entwurf_v2behavior is unchanged and delivery still never imports launch. Fresh call learns a new sibling's garden id from the CALLBACK it makes, never from a lookup — so the pre-injected-token → identity-lookup design (rail §6) is closed, not paused. Do not reopen it without new evidence and explicit GLG re-approval. Fresh-call requires one explicit model and passes it in each runtime's measured CLI dialect (Pi--model <provider/model>, Claude Code--model=<id-or-alias>, Copilot--model <name>beside--interactive <prompt>and the explicit--yolopolicy — a GLG width decision, 2026-08-25, after a callback-only--allow-toolgrant stopped the sibling at every task tool); the Copilot backend opens through entwurf's OWN managed invocation (entwurf copilot), never the bare vendor, and its four required capabilities — birth, MCP hand, receiver, visible footer — are decided bycopilot-fresh-preflight.tsBEFORE the tmux mutation, so a missing unit is a named refusal instead of a dead window; it also takes ONE optional literal absolutecwd(#73 — omitted/""means the caller's own directory; no trim, no realpath, no project-name resolver; classified by the sharedclassify-tmux-cwd.tsleaf; the receipt echoes the REQUEST only), so a cross-repo fresh sibling is a fresh-call fact and never a reason to resume a dormant record. Resume takes only an existing target id and gets transcript/model/provider/cwd from its record. Do not grow either narrow carrier into a generic driver, harness profile, arbitrary command/env selection, or a second creation API. Ownership and import prohibitions: docs/mux-launch-rail.md §11. - One delivery verb:
entwurf_v2addresses an existing garden id. Current routes are live control-socket send, active self-fetch mailbox enqueue, and probe-alive native-push. NO route starts a process: the hidden background resume for a dormant citizen was withdrawn under the visible-first rule, so a dormant in-domain citizen rejects asdormant-fire-forget-unsupportedand the intent axis is single-valued. Every other complementary state pair rejects honestly too. Fresh creation is a separate verb —entwurf_fresh_call— and it returns a launch receipt only; the new sibling's address arrives asynchronously as the sender envelope of its nonce callback.
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.
- today Changed · +2 lines · +196 tokens per session 7f88ac2d8d19
- 4d ago First seen · 223 lines · 7,395 tokens per session scan A a1ee05c2b6a0
entwurf AGENTS.md is an instructions file published in the GitHub repository junghan0611/entwurf (26 stars, last pushed yesterday), licensed MIT. It adds 7,591 tokens to every session, about $0.0380 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 2026-08-30.
Other instructions, from other repositories
intelligent-terminal agent-skills.instructions.md
Guidelines for creating high-quality Agent Skills for GitHub Copilot.
intelligent-terminal rust-localization.instructions.md
Instructions for microsoft/intelligent-terminal, covering localization instructions for rust wta (yaml locale files), file structure, file format, the en-us.yml source file and non-translatable token rules.
intelligent-terminal localization.instructions.md
Instructions for microsoft/intelligent-terminal, covering localization instructions for .resw resource files, components with localized resources, locale categories, authoritative locale list (from terminalapp) and adding or updating localized strings.
intelligent-terminal copilot-instructions.md
Copilot instructions for microsoft/intelligent-terminal, covering copilot instructions for windows terminal, build, running, test and code formatting.
intelligent-terminal instructions.instructions.md
Guidelines for creating high-quality custom instruction files for GitHub Copilot.
intelligent-terminal AGENTS.md
AGENTS.md instructions for microsoft/intelligent-terminal, covering intelligent terminal, architecture, supported agents and settings, user-facing behavior and key files.