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/aaryansinha16/maestro/agents-mdgit clone --depth 1 https://github.com/aaryansinha16/MaestroWrote 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/aaryansinha16/maestro/agents-md)<a href="https://agentmods.dev/instructions/aaryansinha16/maestro/agents-md"><img src="https://agentmods.dev/badge/instructions/aaryansinha16/maestro/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 5.1 | $0.00967 | $0.00967 |
| Opus 5 | $0.00483 | $0.00483 |
| Sonnet 5 | $0.00193 | $0.00193 |
| Haiku 4.5 | $0.00097 | $0.00097 |
Grade A, and why
Maestro 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 5d 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 — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Maestro — AGENTS.md
Purpose
Patterns for AI coding agents working on the Maestro codebase itself. Ensures consistent behavior across sessions.
This file is for agents building Maestro. The system prompts that drive agents working on managed projects are in packages/shared/src/prompt-templates.ts.
Before Any Task
- Read
CLAUDE.mdfor architecture context - Read
DECISIONS.mdfor recent architectural decisions - Read
PROJECT_CONFIG.mdfor the developer's specific projects and their autonomy decisions - Check the relevant package's
src/directory before creating new files - If modifying an existing file, read it fully first
Task Patterns
When adding a new feature to the conductor
- Define the types in
packages/shared/src/types.tsfirst - Add Zod schemas in the same file
- Add the database table/migration in
packages/conductor/src/db.ts - Implement the conductor logic
- Add the API endpoint in
packages/api/src/routes/ - Add the UI integration in
packages/dashboard/src/ - Test end-to-end before considering done
When modifying prompt templates
This is the most sensitive part of the system. Changes here directly affect agent behavior on managed projects.
- Read the existing template fully
- Document the change reasoning in DECISIONS.md
- Test with a "dry run" mode that prints the prompt without running Claude
- Verify on at least one real project before deploying
- Add a version number to the template
When working with the .maestro/ file format
- Always validate with Zod before reading
- If schema changes, write a migration (parse old format, save in new format)
- Never assume the file exists — handle missing gracefully
- Lock files during writes (concurrent sessions could corrupt state)
When adding a new quality gate
- Add the gate type to
packages/shared/src/types.ts - Implement the runner in
packages/conductor/src/quality-gates.ts - Default to non-strict (warn, don't fail) until proven reliable
- Document the gate in
docs/PROMPT_DESIGN.md
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.
- 5d ago First seen · 112 lines · 967 tokens per session scan A 5ee944b25add
Maestro AGENTS.md is an instructions file published in the GitHub repository aaryansinha16/Maestro (2 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 967 tokens to every session, about $0.0048 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
Opencode-Ultrathinker AGENTS.md
AGENTS.md instructions for Kacper0199/Opencode-Ultrathinker, covering global workspace constitution, 1. language & communication, 2. code standards (no comments rule), 3. persistent memory (macrodata) and 4. git protocol & environment.
slipway AGENTS.md
AGENTS.md instructions for signalridge/slipway, covering slipway agent principles, user owns the process, investigate before clarifying, report, do not certify and maintain the whole surface.
slipway CLAUDE.md
Claude Code instructions for signalridge/slipway: Read AGENTS.md. It is the single principle file for this repository, and it applies to Claude Code exactly as written.
factory CLAUDE.md
Claude Code instructions for watt-mind/factory: Claude, please refer to AGENTS.md and docs/protocol.md for the primary repository context, architecture, commands, operational rules, and tracker issue management standards.
RepoResident AGENTS.md
Instructions for ychamel/RepoResident: Agent entry point: read CLAUDE.md in full — it is the operating manual — then follow its Session protocol. Everything else (state, workflows, project knowledge) is routed from there via the .agent/ directory.
factory AGENTS.md
AGENTS.md instructions for watt-mind/factory, covering commands, agent operating floor, handoff, never auto-merge and protected branches.