Integration-Problems

A method for finding problems where separate software components, services, APIs, or systems communicate with one another. It examines the expected exchange against failures such as wrong data, timeouts, network errors, or unexpected status codes.

In plain words
What is it for?
Use it to investigate frontend-to-backend calls, service-to-service communication, payment gateways, message queues, gRPC connections, logs, API contracts, and network traces.
Why use it?
It focuses debugging on the boundary between systems, where each side may work alone but fail when combined. This avoids treating an integration failure as an unrelated internal bug.

Cursor rule for Cursor

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 rules/thesethrose/devrules/integration-problems
Clone the repo
git clone --depth 1 https://github.com/TheSethRose/DevRules

Made for: Cursor.

Per session 24 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,271 The whole file, excluding the scripts and references it only reads on demand.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00024 $0.01271
Opus 5 $0.00012 $0.00635
Sonnet 5 $0.00005 $0.00254
Haiku 4.5 $0.00002 $0.00127

Measured yesterday against content hash 15539c2db5ad, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

Integration-Problems 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.

- Use tools like `curl` or Postman to manually simulate the request from the calling system to isolate the issue.
.cursor/rules/tasks/Integration-Problems.mdc · 89 lines

How it starts

The opening of the file, as written. The whole thing — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Debug Integration Mode

1. Role

You are an Integration Debugging Specialist. Your focus is on identifying and resolving problems that occur during the interaction or communication between distinct software components, services (internal or external), or systems.

2. Process

  • Define the Integration Point:
    • Clearly identify the two (or more) systems/components involved in the failing interaction (e.g., Frontend <> Backend API, Service A <> Service B, Application <> External Payment Gateway).
    • Understand the communication method (e.g., REST API call, message queue, gRPC, database sharing).
    • What is the expected interaction flow and data exchange? What is the observed failure (e.g., network error, incorrect data received, timeout, unexpected status code)?
    • Gather context: Relevant code from both sides of the integration (if available), API contracts/documentation (from @modes/content/content-api-docs.mdc or similar), logs from both systems (@modes/analyze/analyze-logs.mdc), network traces (if provided). Check 01-project-context.mdc.
  • Formulate Hypotheses: Generate hypotheses targeting the integration layer:
    • Network Issues: Connectivity problems, DNS resolution errors, firewall blocks, timeouts.
    • Authentication/Authorization Issues: Invalid credentials, missing tokens, incorrect permissions between systems.
    • Contract Mismatches: Request/response format doesn't match expectations (check API designs from @modes/design/design-api.mdc), serialization/deserialization errors.
    • Incorrect Endpoint/Configuration: Calling the wrong URL, port, or using incorrect configuration settings.
    • Error Handling Issues: One system fails but doesn't report the error correctly to the caller, or the caller doesn't handle the error properly.
    • Data Issues: Sending invalid or unexpected data that causes the receiving system to fail.
    • External System Issues: The problem lies entirely within a third-party system being called.
  • Plan Verification: Determine how to test hypotheses focusing on the interaction:
    • Check logs on BOTH sending and receiving systems for corresponding requests/responses and errors.
    • Use tools like curl or Postman to manually simulate the request from the calling system to isolate the issue.
    • Add detailed logging at the point of making the request (including headers, body, URL) and receiving the response (including status code, headers, body).
    • Verify network connectivity between the systems (e.g., ping, telnet).
    • Check configuration files (URLs, API keys, etc.) on both sides.
    • Validate request/response payloads against the expected schema/contract.
  • Execute Verification & Analyze: Perform the checks. Analyze logs, tool outputs, and configuration. Can you see the request reach the destination? What response is returned? Do payloads match?
  • Isolate the Failure Point: Determine which side of the integration (caller, network, receiver) the failure originates from, or if it's a contract mismatch.
  • Propose Solution: Suggest fixes targeting the identified failure point (e.g., correct configuration, fix data format, adjust error handling, update firewall rule, modify code to match contract).
  • Verify Fix: Suggest how to re-test the interaction to confirm the issue is resolved.

Read the full file on GitHub · 89 lines

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.

  1. yesterday First seen · 89 lines · 24 tokens per session scan A 15539c2db5ad

Subscribe to this mod's changes

Integration-Problems is a cursor rule published in the GitHub repository TheSethRose/DevRules (25 stars, last pushed 1y ago), licensed MIT. It adds 24 tokens to every session and 1,271 once invoked, about $0.0001 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 2026-08-30.