What is sfcode AGENTS.md?

A project guide for sfcode, a command-line coding agent written in PHP with Symfony. It documents the software stack, setup commands, coding rules, and quality checks.

It helps developers install and run sfcode, add tools, follow its PHP and Symfony structure, and run code formatting and static analysis checks.

It gives contributors one place to find the commands and conventions needed to work on the project consistently. It also helps catch style and type errors before changes are considered ready.

sfcode AGENTS.md — Instructions file by yoanbernabeu

Install

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.

agentmods

npx agentmods add instructions/yoanbernabeu/sfcode/agents-md

Clone the repo

git clone --depth 1 https://github.com/yoanbernabeu/sfcode

Made for: Codex, OpenCode.

Per session 465 This file is loaded in full into every session.

When invoked 465 The same file — it is already loaded in full.

Security scan A 1 finding. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00465	$0.00465
Opus 5	$0.00233	$0.00233
Sonnet 5	$0.00093	$0.00093
Haiku 4.5	$0.00047	$0.00047

Measured yesterday against content hash 727066c274f2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

sfcode AGENTS.md scanned grade A with 1 finding 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- PHP 8.4.1+, ext-ctype / curl / iconv / json / mbstring

AGENTS.md · 53 lines

What it actually says

sfcode

CLI coding agent in the spirit of Claude Code, written in PHP with Symfony AI and the Symfony TUI component.

Stack

PHP 8.4.1+, ext-ctype / curl / iconv / json / mbstring
Symfony 8.1 (Console, DependencyInjection, Tui, Runtime)
Symfony AI Platform + Agent (Toolbox, #[AsTool])
amphp/http-client for true async HTTP streaming
Composer 2.x

Running the project

composer install
cp .env.example .env.local && $EDITOR .env.local   # paste OPENAI_API_KEY
php bin/sfcode

Code conventions

PSR-12, declare(strict_types=1); everywhere.
Standard Symfony 8.1 DI kernel (AbstractKernel + KernelTrait + ConsoleBundle).
One tool = one final class with __invoke() and #[AsTool('name', 'description')].
No obvious comments — only the non-trivial why gets a comment.
All code, comments and user-facing strings are in English (the assistant's reply language follows the user's prompt language).
Custom exceptions extend Sfcode\Exception\ExceptionInterface.

Quality gates

composer qa     # php-cs-fixer dry-run + phpstan level 8
composer cs-fix # apply style fixes

Entry points

bin/sfcode → Sfcode\Command\ChatCommand (default command)
Agentic loop core: Sfcode\Agent\AgentLoop::turn() / turnStream()
Tools: src/Tool/*.php (one file = one tool)
Permissions: src/Permission/PermissionManager.php
Skills: src/Skill/SkillLoader.php (project > user cascade, Claude Code compatible)
TUI choreography: src/Tui/TurnRunner.php

Out of scope (by design)

MCP, subagents, hooks, OS sandbox, advanced context compaction, cross-session memory. These are explicitly out of the current scope but documented as future work in CHANGELOG.md.

Changes

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 · 53 lines · 465 tokens per session scan A 727066c274f2

Subscribe to this mod's changes

sfcode AGENTS.md is an instructions file published in the GitHub repository yoanbernabeu/sfcode (5 stars, last pushed 3mo ago), licensed MIT. It adds 465 tokens to every session, about $0.0023 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Languages

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,182 tokens

buildNext

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

sfcode AGENTS.md