Neuron AI is a PHP framework for building AI applications in which agents connect language models, tools, data loaders, vector databases, memory, and user interfaces. PHP developers use it to create and manage applications with agent workflows, multi-agent coordination, streaming, monitoring, and human involvement.
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 skills add neuron-core/neuron-ai --skill neuron-symfony-integrationgit clone --depth 1 https://github.com/neuron-core/neuron-aiWrote 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/skills/neuron-core/neuron-ai/neuron-symfony-integration)<a href="https://agentmods.dev/skills/neuron-core/neuron-ai/neuron-symfony-integration"><img src="https://agentmods.dev/badge/skills/neuron-core/neuron-ai/neuron-symfony-integration/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/skills/neuron-core/neuron-ai/neuron-symfony-integration"><img src="https://agentmods.dev/badge/skills/neuron-core/neuron-ai/neuron-symfony-integration.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 | $0.00221 | $0.14726 |
| Opus 5.5 | $0.00088 | $0.05890 |
| Sonnet 5.5 | $0.00044 | $0.02945 |
| Haiku 4.5 | $0.00022 | $0.01473 |
Grade A, and why
neuron-symfony-integration 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 8d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- `StreamedResponse` takes the generator as chunks and flushes each one: frames reached `curl` one at a time, as `text/event-stream; charset=UTF-8` with `X-Accel-Buffering: no`. `frames()` ends a failure after the first How it starts
The opening of the file, as written. The whole thing — 803 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Neuron AI Symfony Integration
This skill wires Neuron into a Symfony 8 application: first the foundations every feature stands on (packages, layout, keys, tables, container wiring, error mapping, the evaluation command), then the features users ask for, from the agent and the controllers that run it inside a request to the Messenger handler that runs it in a worker, the Stop button, observability, evaluation and tests. Every snippet comes from a reference app (Symfony 8.1, Doctrine DBAL 4, Messenger on Redis) that ran on SQLite, MySQL 8.4, MariaDB 11.7 and PostgreSQL 17. What the agent, the protocols, the approval flow and evaluators mean is owned by the skills under Related; this one owns the Symfony shape.
The Mental Model
The container holds definitions; every request or message works on a bound copy. The agent is a shared service. A controller or handler calls $this->agent->for($threadId), which returns a copy bound to the thread and leaves the shared agent unbound. Per-request pieces (stream adapter, channel, frontend tools) go on the copy. Never call setThreadId() on the injected agent: the second thread served by the same container (the next message in a worker, the next request under a worker runtime or a test client that does not reboot) throws WorkflowException: This workflow is bound to 't-a' and cannot be re-pointed to 't-b'.
Hooks run once per copy, so anything holding a connection is resolved there. The agent pulls its stores through service closures of shared: false services built on Doctrine's current PDO. The shared agent never resolves them; each copy gets its own instances.
| Piece | Where | Lifetime |
|---|---|---|
SupportAgent: provider hook, instructions, tools() hook |
src/Neuron/Agents, autowired |
Shared service, never bound |
| Bound copy + adapter, channel, frontend tools | Controller action, message handler | One request or one message |
| Tools, scoped to the thread's customer | Built by tools() on each copy |
One copy |
MessageStoreInterface, PersistenceInterface, neuron.pdo |
config/packages/neuron.yaml |
shared: false: new per resolution, current PDO |
WorkflowEngine |
Autowired, shared: false |
Reads runs without building the agent |
Provider wrapped in StoppableHttpClient |
Agent provider() hook |
Built for every execution segment |
| Stop flags | Cache pool neuron.stop_signals |
Raised by Stop, cleared by the next turn (5 minutes at most) |
| Thread ownership | ThreadVoter + #[IsGranted] |
Every Neuron route |
| Neuron exceptions to statuses | NeuronExceptionListener |
kernel.exception |
| Background turn | RunSupportAgent message + handler |
Run ID minted at dispatch |
| Evaluators | src/Neuron/Evaluators, run by bin/console neuron:evaluate |
One copy with in-memory stores per dataset item |
What ships with it
6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 8d ago Changed 6e003b0ce0fd
- 11d ago Changed · +2 lines 5ade2a8402f9
- 12d ago Changed · +6 lines 456c7012793a
- 15d ago First seen · 795 lines · 221 tokens per session scan A a91cf4f4ed16
neuron-symfony-integration is a skill published in the GitHub repository neuron-core/neuron-ai (2,140 stars, last pushed yesterday), licensed MIT. It adds 221 tokens to every session and 14,726 once invoked, about $0.0009 per session on Opus 5.5. 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 2026-10-01.
Other skills, from other repositories
neuron-agent
Create, configure, and extend Neuron AI agents in PHP using providers, tools, chat history, memory, and custom workflow nodes. Use for Neuron Agent implementation, configuration, conversation lifecycle, and input/output extensions.
laravel
Use when building or extending a Laravel 11/12 app — Eloquent models, migrations and relationships, routing with controllers and Form Requests, queues and background jobs, framework-native security (validation, mass-assignment, policies, signed URLs, rate limiting), and Pest/PHPUnit feature tests. NOT pure PHP…
php-laravel
Build and upgrade PHP/Laravel applications with validated requests, policy authorization and reliable jobs/data access. Use when implementing Laravel routes, Eloquent models or queues; preserve existing project conventions and runtime compatibility.
scraperapi-php-sdk
Best-practices reference for the ScraperAPI PHP SDK (scraperapi/sdk Composer package). Consult whenever the user is writing, debugging, or reviewing PHP code that calls ScraperAPI. Use when user asks: "scrape a website with PHP and ScraperAPI", "ScraperAPI PHP example", "how do I use the ScraperAPI PHP SDK", "PHP…
pn-php-scaffolding
Scaffolds new PHP API projects (Laravel, Slim) or routes. Use when adding a new controller/service; covers PSR standards, Composer hygiene, Laravel conventions, and idiomatic PHP 8.x patterns.
artisan-commands
Execute Laravel Artisan commands safely.