Borrowing it
Nothing to install: this file belongs to anistark/sutras. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/anistark/sutras/main/AGENTS.mdgit clone --depth 1 https://github.com/anistark/sutrasWrote 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/anistark/sutras/agents-md)<a href="https://agentmods.dev/instructions/anistark/sutras/agents-md"><img src="https://agentmods.dev/badge/instructions/anistark/sutras/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/anistark/sutras/agents-md"><img src="https://agentmods.dev/badge/instructions/anistark/sutras/agents-md.svg" alt="Reviewed on agentmods" width="80" 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 | $0.03648 | $0.03648 |
| Opus 5.5 | $0.01459 | $0.01459 |
| Sonnet 5.5 | $0.00730 | $0.00730 |
| Haiku 4.5 | $0.00365 | $0.00365 |
Grade A, and why
sutras 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 26d 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 — 322 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — AI Coding Agent Instructions for Sutras
Instructions for Claude Code, Cursor, Copilot, pi, opencode and other AI coding agents working on this project.
Project Overview
Sutras is the universal devtool for agent skills — create, test, evaluate, distribute, and discover skills for AI coding agents (Claude Code, Cursor, Windsurf, Copilot, and more).
- Repository: https://github.com/anistark/sutras
- PyPI: https://pypi.org/project/sutras/
- Docs: https://anistark.github.io/sutras/
- License: MIT
- Python: >=3.11
- Build system: hatchling (via
uv)
Architecture
Sutras is a Python CLI built with Click. It has two layers:
1. Core Library (src/sutras/core/)
The engine — models, loaders, runners, and managers. All business logic lives here.
| File | Purpose |
|---|---|
abi.py |
Pydantic models for sutras.yaml (SutrasABI, TestConfig, EvalConfig, etc.) |
skill.py |
Skill model — combines SKILL.md frontmatter with ABI metadata |
loader.py |
SkillLoader — discovers and loads skills from ~/.claude/skills/ and .claude/skills/ |
config.py |
SutrasConfig — global config from ~/.sutras/config.yaml, registry entries |
templates.py |
Built-in skill templates for sutras new --template |
test_runner.py |
TestRunner — executes test cases defined in sutras.yaml |
evaluator.py |
Evaluator — Ragas-based evaluation framework |
builder.py |
SkillBuilder — packages skills into tarballs for distribution |
naming.py |
SkillName parser — handles @namespace/skill and bare names |
semver.py |
Semver constraint parsing (npm-style: ^1.0.0, ~1.2.3, ranges) |
registry.py |
RegistryManager — federated git-based registry system |
installer.py |
SkillInstaller — installs skills to ~/.claude/installed/ with symlinks |
publisher.py |
SkillPublisher — publishes skills to registries (direct push or PR) |
resolver.py |
DependencyResolver — recursive dependency resolution with conflict detection |
lockfile.py |
.sutras.lock format — lock file for reproducible installs |
docgen.py |
Documentation generation from skill metadata |
validation.py |
Limits, known tool names, and link extraction used by sutras validate |
bench.py |
BenchPlan, BenchRunner (matrix runs, spend cap), BenchReport, history, --record compatibility |
runtime/ |
RuntimeExecutor protocol and ClaudeCodeExecutor (headless claude -p, stream-json parsing, model discovery) |
sandbox.py |
Temp workspaces with the skill installed at project level; workspace diffs |
assertions.py |
Deterministic bench assertions |
judge.py |
Rubric grading with a fixed judge model |
pricing.py |
Price table, model aliases, model@effort parsing |
estimator.py |
Pre-run bench cost estimates (history, pilot, heuristic) |
updater.py |
Self-update logic — checks PyPI, upgrades CLI, refreshes pi extension |
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.
- 26d ago Changed · +1 lines · +44 tokens per session b875d535d22e
- 1mo ago Changed · +10 lines · +302 tokens per session e8e6f37ac112
- 1mo ago Changed · +1 lines · +23 tokens per session 7335ed6faceb
- 2mo ago First seen · 310 lines · 3,279 tokens per session scan A b8887051ccb1
sutras AGENTS.md is an instructions file published in the GitHub repository anistark/sutras (5 stars, last pushed 17d ago), licensed MIT. It adds 3,648 tokens to every session, about $0.0146 per session on Opus 5.5. 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.
sutras AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| sutras AGENTS.mdanistark/sutras | Instructions file | 5 | 17d ago | MIT | A | 3,648 | Codex, OpenCode |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | yesterday | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | today | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 275k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |