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/gabrielmaialva33/winx-code-agent/agents-mdgit clone --depth 1 https://github.com/gabrielmaialva33/winx-code-agentWrote 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/gabrielmaialva33/winx-code-agent/agents-md)<a href="https://agentmods.dev/instructions/gabrielmaialva33/winx-code-agent/agents-md"><img src="https://agentmods.dev/badge/instructions/gabrielmaialva33/winx-code-agent/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 | $0.00868 | $0.00868 |
| Opus 5 | $0.00434 | $0.00434 |
| Sonnet 5 | $0.00174 | $0.00174 |
| Haiku 4.5 | $0.00087 | $0.00087 |
Grade A, and why
winx-code-agent 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 4d 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 — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository Guidelines
Project Structure & Module Organization
Winx is a Rust 2021 MCP server for shell and coding-agent workflows. The binary entry point is src/main.rs; library exports live in src/lib.rs. src/server.rs is the MCP-service facade, with catalog/schema, handler, principal isolation, session, Task, dispatch, and tests under src/server/; hardened HTTP transport lives in src/http_server.rs, and typed secret/config parsing in src/config.rs. Tool facades live under src/tools/; BashCommand and FileWriteOrEdit are split into focused submodules. Terminal and persistence logic belongs under src/state/, reusable parsing/path/repository/safety helpers under src/utils/, and audited libc calls exclusively under src/os.rs (the crate otherwise denies unsafe code). Integration tests live in tests/, Criterion benchmarks in benches/, cargo-fuzz targets in fuzz/, model assets in assets/, and pinned CI/release automation in .github/workflows/.
Build, Test, and Development Commands
cargo check --all-features --lockedperforms a fast compile check with the committed lockfile.WINX_EMBEDDED=1 cargo run --releasestarts an in-process development server without sibling daemons.cargo build --release --locked --binsproduces the complete optimized Unix bundle; run./target/release/winx-code-agentafterward.cargo fmt --all -- --checkverifies formatting; runcargo fmt --allto apply it.cargo clippy --all-targets --all-features --locked -- -D warningsruns the enforced lint suite.cargo test --all-features --lockedruns the full normal test suite.cargo test --features loom --lib loom_runs the opt-in concurrency model checks.cargo +1.88.0 check --all-features --lockedverifies the declared MSRV.cargo package --lockedverifies the crates.io package.cargo bench --bench performance --locked --no-runcompiles the benchmark suite.cargo deny --all-features checkandcargo audit --deny warningsenforce dependency policy.cargo +nightly fuzz buildcompiles all cargo-fuzz targets.
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.
- 4d ago First seen · 37 lines · 868 tokens per session scan A c8d9308e90a3
winx-code-agent AGENTS.md is an instructions file published in the GitHub repository gabrielmaialva33/winx-code-agent (33 stars, last pushed 3d ago), licensed MIT. It adds 868 tokens to every session, about $0.0043 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
flow-next CLAUDE.md
Claude Code instructions for gmickel/flow-next, covering claude code project guide, stack and tooling, when to use a skill (the default), when to use deterministic flowctl python and the common pattern: skill + thin flowctl plumbing.
skyvern CLAUDE.md
Claude Code instructions for Skyvern-AI/skyvern, covering claude.md, development commands, python backend commands, code quality & testing and frontend commands (in skyvern-frontend/).
OwnPilot AGENTS.md
Instructions for ownpilot/OwnPilot, covering ownpilot, architecture, key patterns, commands and tech stack.
MyAgents AGENTS.md
AGENTS.md instructions for hAcKlyc/MyAgents, a project described as: MyAgents - 优雅、易用的 Agent 桌面端 ，一站式 Agent 工作台与任务中心.
AutoResearch AGENTS.md
Instructions for Maleick/AutoResearch, covering auto research agent guide, project purpose, source of truth, development rules and security rules.
agent-computer-use CLAUDE.md
Instructions for kortix-ai/agent-computer-use, covering agent-computer-use, repo structure, quick reference, working on the cli and working on docs.