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/team-reflect/reflect-open/agents-mdgit clone --depth 1 https://github.com/team-reflect/reflect-openWhat 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 | $0.04201 | $0.04201 |
| Opus 5 | $0.02100 | $0.02100 |
| Sonnet 5 | $0.00840 | $0.00840 |
| Haiku 4.5 | $0.00420 | $0.00420 |
Grade A, and why
reflect-open 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.
How it starts
The opening of the file, as written. The whole thing — 308 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Purpose
This document helps AI agents and automated systems interact with the Reflect repo safely and effectively. It summarizes setup, workflows, CI parity, testing, directories, and environment variables.
What is Reflect
Reflect is a modern note‑taking tool with a TypeScript codebase. This repo contains Reflect V2, a rewrite of the original Reflect code-base to make it offline-first, markdown backed, and open source.
Product Principles
Drawn from the product docs — read these for deeper context: V1 Overview · V2 Product Vision · V2 Grounding Brief · Indexing Strategy · Sync Strategy
- Daily notes first. The app opens to today's note. All capture flows into the daily note by default.
- Association over hierarchy.
[[Wiki Links]]replace folders. The note graph is the organizing model; there are no folders. - Markdown is the source of truth. Notes are
.mdfiles (daily/YYYY-MM-DD.md,notes/). SQLite under.reflect/is a rebuildable projection of the notes — with one durable exception: thechat_*tables hold AI chat history, which is not derivable from markdown. Index wipes and rebuilds must leave them untouched. - No Reflect-hosted APIs. LLM calls go directly to user-approved providers (OpenAI, Anthropic, etc.). Sync goes to GitHub/iCloud/Git. Never proxy through Reflect infrastructure.
- BYOK AI. AI features use user-supplied keys. Never assume Reflect operates AI infrastructure.
private: trueis a hard block. Notes with this frontmatter flag must never have their content sent to any external service — AI, transcription, or otherwise. Enforce at every call site.- Keyboard-native UX. Every core workflow must be reachable from the keyboard. This is product identity, not polish.
- Minimal UI. Do less, and do it well. Don't add surfaces that compete with the editor.
- Secrets in the OS keychain. API keys and credentials never go in markdown, Git, or
.reflect/. - Portable data. Full export (JSON, markdown, HTML) must work from day one.
- No Electron. Desktop shell is Tauri.
- MIT open-source core. Write as if the code is public and will be critiqued.
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 · 308 lines · 4,201 tokens per session scan A 7b03618c8c8f
reflect-open AGENTS.md is an instructions file published in the GitHub repository team-reflect/reflect-open (1,445 stars, last pushed yesterday), licensed MIT. It adds 4,201 tokens to every session, about $0.0210 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
eve AGENTS.md
Instructions for vercel/eve, covering agents.md, about eve, repository layout, git workflow and commands.
Backlog.md AGENTS.md
Instructions for MrLesk/Backlog.md, covering read this first, simplicity-first implementation rules, ui descriptions, commands and development.
Backlog.md copilot-instructions.md
Instructions for MrLesk/Backlog.md, covering ⚠️ important, commands, development, testing and configuration management.
siyuan AGENTS.md
AGENTS.md instructions for siyuan-note/siyuan, covering agents.md, 1. non-negotiable constraints, do not hand-edit, verification and prohibited operations and 2. project-specific rules.
shiki CLAUDE.md
Instructions for sazardev/shiki, covering claude.md, project, commands, versioning and marketing site (/docs, github pages).
shiki AGENTS.md
Instructions for sazardev/shiki, covering agents.md, read this first, commands, testing and invariants that are easy to break.