This specification is maintained by the AgentRules Working Group. Have enhancements or want to claim maintainership?Submit Specification →
backendcritical PriorityID: axum-0-8-state-tower-panic-resilience

Axum 0.8 Type-Safe State Extractors, Tower Middleware Ordering, and Panic Boundary Architecture

Mandates compile-time type-safe state injection via Axum 0.8 State and FromRef extractors, strictly forbidding untyped Extension usage. Enforces deterministic Tower ServiceBuilder layer execution order and requires CatchPanicLayer boundaries to guarantee zero connection-dropping unhandled panics.

Prohibited Anti-Pattern: Dynamic runtime Extension extraction leading to unhandled runtime panics, inverted Tower middleware evaluation, and naked worker panics dropping HTTP sockets.

Export Ready-to-Drop Configuration

Select format below to copy code or download file directly.

---
description: Axum 0.8 Type-Safe State Extractors, Tower Middleware Ordering, and Panic Boundary Architecture - Mandates compile-time type-safe state injection via Axum 0.8 State and FromRef extractors, strictly forbidding untyped Extension usage. Enforces deterministic Tower ServiceBuilder layer execution order and requires CatchPanicLayer boundaries to guarantee zero connection-dropping unhandled panics.
globs: src/**/*.rs, crates/**/*.rs
alwaysApply: true
verified_by: "agentrules.org"
license: "CC-BY-4.0"
source: "https://agentrules.org/rules/axum-0-8-state-tower-panic-resilience"
---

# Axum 0.8 Type-Safe State Extractors, Tower Middleware Ordering, and Panic Boundary Architecture

> **Severity**: CRITICAL | **Category**: backend
> **Anti-Pattern Warning**: Dynamic runtime Extension extraction leading to unhandled runtime panics, inverted Tower middleware evaluation, and naked worker panics dropping HTTP sockets.

## System Constraints
You are a Principal Rust Systems Architect enforcing production-grade Axum 0.8 and Tower invariants. Mandate compile-time type checking with State extractors, explicit sub-state derivation via FromRef, ordered Tower ServiceBuilder layers, and comprehensive CatchPanicLayer isolation to guarantee HTTP 500 responses without dropped client connections.

## Strict Guidelines
- Engineers MUST ALWAYS use axum::extract::State with axum::extract::FromRef for dependency injection instead of dynamic Request Extensions to catch missing state dependencies at compile time.
- Engineers MUST NOT use Extension<T> for application state, database connection pools, or core business clients; Extension is exclusively reserved for transient per-request middleware context.
- Services MUST wrap all outer router pipelines with tower_http::catch_panic::CatchPanicLayer configured with a custom panic response builder returning structured application/json errors (HTTP 500).
- Services MUST compose middleware using tower::ServiceBuilder to ensure correct LIFO/FIFO evaluation order, placing tracing, request metrics, and panic handling at the outermost boundary.
- Handlers MUST ALWAYS place body-consuming extractors (such as Json, Form, or Bytes) as the final parameter in the handler argument list to satisfy Axum's single-pass body consumption invariant.
- Engineers MUST implement axum::response::IntoResponse for all domain error types using explicit enum variants, prohibiting String or anyhow::Error from bubbling directly to handlers.

## Bad vs Good Implementations
### Axum State & Tower Middleware Pipeline

#### Bad Antipattern (AI Common Hallucination)
> **Why it fails**: Using Extension<T> bypasses Rust's compile-time type-checker; omitting the extension causes a 500 internal server error or unhandled panic during request evaluation. Extracting Json before other parameters can lead to subtle parser issues, and absent CatchPanicLayer aborts the TCP stream on panics.
```rust
use axum::{routing::post, Extension, Json, Router};
use std::sync::Arc;

// ANTI-PATTERN: Untyped runtime Extension extraction and missing panic boundaries
#[derive(Clone)]
struct AppState { db: Arc<String> }

// If Extension<Arc<String>> is omitted from router, this panics at runtime during request handling!
// Placing Json<Payload> before headers/state violates single-pass extractor convention
async fn handler(
    Json(body): Json<serde_json::Value>,
    Extension(db): Extension<Arc<String>>,
) -> String {
    if body.get("invalid").is_some() {
        panic!("Critical failure!"); // Drops connection abruptly without HTTP response
    }
    format!("Processed with: {}", db)
}

fn make_router() -> Router {
    Router::new()
        .route("/api/resource", post(handler))
        // Missing State provider, missing CatchPanicLayer, missing TraceLayer
}
```

#### Production Standard (Correct)
> **Standard**: Leverages Axum 0.8 State with FromRef for compile-time verified sub-state injection. Arranges the body-consuming Json extractor as the final parameter, wraps the service with ServiceBuilder maintaining correct layer execution, and handles thread panics cleanly via CatchPanicLayer.
```rust
use axum::{
    extract::{FromRef, State},
    http::StatusCode,
    response::{IntoResponse, Response},
    routing::post,
    Json,
    Router,
};
use serde::Deserialize;
use std::sync::Arc;
use tower::ServiceBuilder;
use tower_http::catch_panic::CatchPanicLayer;
use tower_http::trace::TraceLayer;

#[derive(Clone)]
pub struct AppState {
    pub db_pool: Arc<String>,
    pub config_val: Arc<String>,
}

// Implement FromRef to allow granular, decoupled state extraction
impl FromRef<AppState> for Arc<String> {
    fn from_ref(state: &AppState) -> Self {
        state.db_pool.clone()
    }
}

#[derive(Deserialize)]
pub struct RequestPayload {
    pub data: String,
}

// GOOD: State checked at compile-time; body-consuming Json extractor is the LAST parameter
async fn handler(
    State(db): State<Arc<String>>,
    Json(payload): Json<RequestPayload>,
) -> Result<impl IntoResponse, AppError> {
    Ok((StatusCode.OK, Json(serde_json::json!({"status": "ok", "db": *db}))))
}

enum AppError {
    Internal,
}

impl IntoResponse for AppError {
    fn into_response(self) -> Response {
        (StatusCode::INTERNAL_SERVER_ERROR, "Internal Error").into_response()
    }
}

fn handle_panic(_err: Box<dyn std::any::Any + Send + 'static>) -> Response {
    (StatusCode::INTERNAL_SERVER_ERROR, Json(serde_json::json!({"error": "Panic occurred"}))).into_response()
}

pub fn build_router(state: AppState) -> Router {
    let middleware_stack = ServiceBuilder::new()
        .layer(TraceLayer::new_for_http())
        .layer(CatchPanicLayer::custom(handle_panic));

    Router::new()
        .route("/api/resource", post(handler))
        .layer(middleware_stack)
        .with_state(state)
}
```

Architectural Specification & Invariant Rationale

Architectural Invariant Design: Axum 0.8 Type-Safe State Extractors, Tower Middleware Ordering, and Panic Boundary Architecture

Axum 0.8 Type-Safe State Extractors, Tower Middleware Ordering, and Panic Boundary Architecture defines an industrial-grade behavioral standard for backend engineering. Mandates compile-time type-safe state injection via Axum 0.8 State and FromRef extractors, strictly forbidding untyped Extension usage. Enforces deterministic Tower ServiceBuilder layer execution order and requires CatchPanicLayer boundaries to guarantee zero connection-dropping unhandled panics. Modern software architectures require strict adherence to deterministic patterns to guarantee high availability, maintainability, and zero security vulnerabilities.

Framework Paradigm Shift & Drift MechanicsLLMs are trained on millions of open-source repositories containing outdated code, legacy idioms, and unpatched antipatterns. When coding assistants generate implementations for this stack, they default to common statistical tokens: Dynamic runtime Extension extraction leading to unhandled runtime panics, inverted Tower middleware evaluation, and naked worker panics dropping HTTP sockets. Without an explicit system instruction file, developers spend hours debugging silent failures.
Failure Modes & Production Blast Radius

Hazard Analysis: What Breaks Without This Rule

The following failure modes have been cataloged from hundreds of hours of production code reviews where AI assistants produced non-compliant implementations:

Failure ModeAI Symptom (The Anti-Pattern)SeverityProduction Impact & Remedy
AI Hallucination & Deprecation DriftDynamic runtime Extension extraction leading to unhandled runtime panics, inverted Tower middleware evaluation, and naked worker panics dropping HTTP sockets.Critical
Violating this industrial standard introduces regression bugs, security risks, and unexpected runtime failures in production.
Enforce explicit behavioral system directives across all AI coding assistants in your repository.
Silent Architecture InconsistencyAI assistants produce mixed paradigms across files, creating fragmented codebase conventions.Medium
Increases cognitive load during code reviews and leads to maintenance debt across engineering teams.
Pin project conventions in .cursor/rules, CLAUDE.md, and AGENTS.md so all team agents follow identical guardrails.
Architectural Directives in Detail

Step-by-Step Production Directives

Each rule constraint below is injected into the AI agent's system prompt to establish deterministic boundaries:

1

Engineers MUST ALWAYS use axum

Engineers MUST ALWAYS use axum::extract::State with axum::extract::FromRef for dependency injection instead of dynamic Request Extensions to catch missing state dependencies at compile time.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
2

Engineers MUST NOT use Extension<T> for application state, database connection pools, or core business clients

Engineers MUST NOT use Extension<T> for application state, database connection pools, or core business clients; Extension is exclusively reserved for transient per-request middleware context.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
3

Services MUST wrap all outer router pipelines with tower_http

Services MUST wrap all outer router pipelines with tower_http::catch_panic::CatchPanicLayer configured with a custom panic response builder returning structured application/json errors (HTTP 500).

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
4

Services MUST compose middleware using tower

Services MUST compose middleware using tower::ServiceBuilder to ensure correct LIFO/FIFO evaluation order, placing tracing, request metrics, and panic handling at the outermost boundary.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
5

Handlers MUST ALWAYS place body-consuming extractors (such as Json, Form, or Bytes) as the final parameter in the handler argument list to satisfy Axum's single-pass body consumption invariant.

Handlers MUST ALWAYS place body-consuming extractors (such as Json, Form, or Bytes) as the final parameter in the handler argument list to satisfy Axum's single-pass body consumption invariant.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
6

Engineers MUST implement axum

Engineers MUST implement axum::response::IntoResponse for all domain error types using explicit enum variants, prohibiting String or anyhow::Error from bubbling directly to handlers.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
Technical Questions & Answers

Frequently Asked Technical Questions

Common questions developers face when configuring AI coding assistants for this stack:

Q:Why is Axum 0.8 Type-Safe State Extractors, Tower Middleware Ordering, and Panic Boundary Architecture critical for AI coding assistants?

Mandates compile-time type-safe state injection via Axum 0.8 State and FromRef extractors, strictly forbidding untyped Extension usage. Enforces deterministic Tower ServiceBuilder layer execution order and requires CatchPanicLayer boundaries to guarantee zero connection-dropping unhandled panics. Without explicit behavioral boundaries, AI coding assistants (Cursor, Claude Code, Windsurf, Codex) rely on training data averages, which frequently produce deprecated, inefficient, or vulnerable code patterns.

Q:How does adopting this rule prevent AI hallucinations?

This rule explicitly declares what the AI must avoid (Dynamic runtime Extension extraction leading to unhandled runtime panics, inverted Tower middleware evaluation, and naked worker panics dropping HTTP sockets.) and prescribes verified positive examples. Grounding the LLM with negative and positive pairs eliminates guesswork.

Q:Which AI coding environments are supported?

This specification is multi-platform compatible: Cursor (.cursor/rules/*.mdc & .cursorrules), Anthropic Claude Code (CLAUDE.md), Windsurf (.windsurfrules), Codex / Cline (AGENTS.md), and GitHub Copilot (.github/copilot-instructions.md).

Q:Can I combine this rule with other technology stacks?

Yes. Use the Interactive Stack Builder or download an Agent Pack to merge this rule with other frontend, backend, database, security, and DevOps rules without syntax collision.

Open Specification · CC BY 4.0·Free for Personal, Team & Commercial In-Repo Use

Citation & Open Source Attribution Guidelines

This specification is licensed under Creative Commons Attribution 4.0 International (CC BY 4.0). You are completely free to adapt and share this rule in your workspace. If citing, including in public AI benchmark datasets, or redistributing, attribution to AgentRules.org is required.

1 Rule in Stack

Axum