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/fastrepl/anarlog/agents-mdgit clone --depth 1 https://github.com/fastrepl/anarlogWhat 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 | $0.02022 | $0.02022 |
| Opus 5 | $0.01011 | $0.01011 |
| Sonnet 5 | $0.00404 | $0.00404 |
| Haiku 4.5 | $0.00202 | $0.00202 |
Grade A, and why
anarlog 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 — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Overview
Tauri desktop note-taking app (apps/desktop/) with a web app (apps/web/).
Uses pnpm workspaces.
SQLite is the primary data store (schema and migrations in crates/db-app/, desktop transport in plugins/db/), Zustand is used for UI state, and ProseMirror powers the editor (packages/editor, via @handlewithcare/react-prosemirror); documents are stored as TipTap-dialect ProseMirror JSON (converters/validation in crates/tiptap). Sessions are the core entity — all notes are backed by sessions.
Commands
- Format:
pnpm exec dprint fmt - Typecheck (TS):
pnpm -r typecheck - Typecheck (Rust):
cargo check - Desktop dev:
turbo dev:desktop - Web dev:
turbo dev:web - Dev docs: https://docs.anarlog.so
Pre-commit verification
- Before every commit, run the locally available checks from the CI workflows affected by the changed paths. Do not defer routine validation to CI.
- Always run
pnpm exec dprint fmt, thenpnpm fmt:check. - For TypeScript changes, run the affected package's typecheck and tests. For desktop changes, run
pnpm -F desktop typecheckandpnpm -F desktop test; usepnpm -r typecheckwhen changes span packages. - For desktop TypeScript changes, run the CI lint command:
pnpm exec oxlint --quiet --format=github apps/desktop/src/. - When desktop translated copy, message extraction, or catalogs may change, run
pnpm -F desktop exec lingui extract --clean --workers 1andpnpm -F desktop exec lingui compile --strict --workers 1, include all generatedapps/desktop/src/i18n/localeschanges, and rerun until generation is stable. When there is no intentional uncommitted catalog diff, run the exact CI check:pnpm -F desktop i18n:check. - For Rust changes, run
cargo checkand the affected package tests. Match stricter workflow commands such ascargo clippy --locked ... -D warningswhen the changed paths trigger them. - Check the relevant
.github/workflows/*_ci.*,fmt.yaml, andlint.yamlfiles for package-specific tests, generated-file checks, or build validation, and run the locally reproducible commands before committing. - If a required check fails, fix it before committing. If a check cannot run locally because of platform, secrets, or unavailable infrastructure, report the exact skipped command and reason; do not claim full validation.
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 · -10 tokens per session 627b215bb0cb
- 2d ago First seen · 86 lines · 2,032 tokens per session scan A 89d31b9527b8
anarlog AGENTS.md is an instructions file published in the GitHub repository fastrepl/anarlog (9,219 stars, last pushed today), licensed MIT. It adds 2,022 tokens to every session, about $0.0101 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
youwee AGENTS.md
Instructions for vanloctech/youwee, covering agents.md — project rules for ai assistants, project info, mandatory checklists, version bump checklist and commit rules.
phase CLAUDE.md
Instructions for phase-rs/phase, covering claude.md, project overview, design principles — read this first, when in doubt and agent team orchestration standards.
Hermes-CN-Desktop AGENTS.md
AGENTS.md instructions for Eynzof/Hermes-CN-Desktop, covering codex 工作指引, 项目概述, 项目结构, 后端事实来源 and 后端版本同步.
Hermes-CN-Desktop CLAUDE.md
Claude Code instructions for Eynzof/Hermes-CN-Desktop, covering claude 工作指引, 项目概述, 项目结构, 后端事实来源 and 开发流程.
asyar AGENTS.md
AGENTS.md instructions for Xoshbin/asyar, covering asyar project agent guidelines & rules, 1. strict git policy (no git writes without asking), 2. no ai attribution, 3. mandatory verification & local ci matrix and 4. formatting enforcement.
Unfour AGENTS.md
Instructions for zyqzyq/Unfour, covering agents.md, product and module scope, package boundary rules, backend and command-bus rules and ai execution rules.