Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
DeepSeek
Muse+22 moreClaude CodeCursorCodex CLIDeepSeekMuse
Dots
Manus
Grok
Grok Bot+18 agentsJS NetworkResilience — Static AST Resilience Auditor
Detects un-timeouted network calls, unguarded fetches, and shared-failure domain fallbacks in Node.js.
$19.99
JS NetworkResilience — Static AST Resilience Auditor
Example session with this skill installed
// Combined input sample showing network resilience and fallback vulnerabilities
const axios = require('axios');
const PRIMARY_HOST = 'https://primary-provider.com';
const FALLBACK_HOST = 'https://primary-provider.com';
async function ask(payload) {
try {
const response = await axios.post(PRIMARY_HOST, payload, { timeout: 5000 });
return response.data;
} catch (primaryError) {
console.warn('primary failed:', primaryError.message);
const backup = await axios.post(FALLBACK_HOST, payload);
return backup.data;
}
}
module.exports = { ask };
- Read your context and instructions
- Compiled the js networkresilience
{
"audit_status": "CRITICAL",
"findings": [
{
"id": "f1",
"file": "live/scan2.js",
"line": 5,
"severity": "HIGH",
"signature": "SHARED_FAILURE_DOMAIN",
"message": "FALLBACK_HOST resolves to the same host as PRIMARY_HOST",
"snippet": "const FALLBACK_HOST = 'https://primary-provider.com';"
},
{
"id": "f2",
"file": "live/scan2.js",
"line": 12,
"severity": "MEDIUM",
"signature": "NON_ISOLATED_LEDGER",
"message": "failover logged via root logger, not a dedicated fallback logger",
"snippet": "console.warn('primary failed:', primaryError.message);"
},
{
"id": "f3",
"file": "live/scan2.js",
"line": 13,
"severity": "HIGH",
"signature": "MISSING_TIMEOUT",
"message": "axios.post inside catch block has no timeout configured",
"snippet": "const backup = await axios.post(FALLBACK_HOST, payload);"
}
],
"metadata": {
"version": "1.0.0",
"timestamp": "2026-10-15T12:00:00Z",
"files_scanned": 1,
"calls_found": 2
}
}
Connects securely to your tools. The creator never sees your data.
About this skill
Stop debugging silent backend hangs, thread pool starvation, and unmitigated third-party provider outages. JS NetworkResilience is a static AST-driven architectural auditor that scans Node.js repositories to expose dangerous network execution paths and fragile failover mock infrastructure before code reaches production staging.
True runtime resilience cannot be achieved by merely wraping network clients in basic, unconfigured try/catch blocks that log failures to generic stdout pipelines. This engine evaluates your active outbound boundaries to enforce deterministic circuit-breaking policies and strict fallback isolation contracts.
What it does
- Outbound Client Interception: Deep-scans files via Abstract Syntax Tree (AST) patterns to isolate unconfigured integration hooks across axios, fetch, got, and node-fetch libraries.
- Resilience Schema Enforcement: Evaluates network clients against six rigid execution signatures, discarding conversational or heuristic guessing models.
- Thread Pool Starvation Guard: Flags every outbound request lacking a deterministic inline timeout layer or AbortController configuration.
- Failover Fault-Zone Isolation: Detects and blocks critical backup configuration flaws where primary and secondary host definitions resolve back to identical infrastructure infrastructure.
- Machine-Readable Telemetry: Outputs a strict, schema-validated JSON payload optimized for local pre-commit hooks, IDE diagnostic buffers, or automated CI/CD code review gates.
Detected Structural Vulnerabilities
- MISSING_TIMEOUT: Executing HTTP requests without explicit configuration parameters or AbortController boundaries, allowing slow network peers to freeze the single-threaded Node event loop indefinitely.
- UNGUARDED_NETWORK_CALL: Outbound calls executed entirely outside of a synchronized try/catch architecture or an active unhandled .catch() promise settlement array.
- MISSING_FALLBACK: Critical functional scripts processing external APIs without an identified module-level backup destination or alternative host proxy mapping.
- SHARED_FAILURE_DOMAIN: Multi-host fallback configurations where the secondary endpoint signature resolves to the exact server architecture zone as the primary gateway, causing secondary connection failure.
- NON_ISOLATED_LEDGER: Routing architectural failover trace logs directly into primitive console logging sinks (console.warn, console.log) instead of isolated, telemetry-monitored ledger managers.
- MISSING_PAYLOAD_CHECK: Foward-routing raw data schemas to secondary validation sinks without explicitly proving the endpoint signatures match across shifting protocol schemas.
When to use
- Auditing highly concurrent microservices, API gateways, or messaging integrations relying on external webhooks.
- Conducting rigid architectural compliance validations across third-party framework layers or vendor plugins.
- Reviewing automated AI-generated infrastructure and client code structures for production-grade resilience standards.
When NOT to use
- Monitoring active runtime environments (this engine functions strictly in a passive, read-only static analysis orientation).
- Codebases outside the native JavaScript and TypeScript backend environments.
- Scanning single-file utility setups or integration prototypes under 300 total lines of code.
Under the Hood & Security
- 100% Local & Air-Gapped: Operates fully within local memory buffers (file_system_write: false). Absolute zero external cloud validation queries ensure sensitive endpoints, internal schemas, and corporate workflows never leak.
- Contract-Tested Stability: Backed by a rigorous 12-point testing harness validating schema compliance boundaries, edge cases, and absolute zero-hallucination execution.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Code Quality & Review.
Static analysis constraints only. Inspects client parameters inside local memory buffers without establishing external host handshakes, executing local node runtimes, or writing automated refactoring patches. Timeouts or abort strategies established inside external shared client instances are not programmatically visible to the parser sequence. Scans are bypassed on lightweight scripts under 300 total lines of code.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 1 month ago
- Software
Node.jsaxios
Permissions
File Scopes
Read-only sandbox access to local JavaScript/TypeScript integration modules. Evaluates client configurations completely offline with zero outbound network calls, telemetry phone-home vectors, or codebase modification routines.