project-architecture

A set of project-specific rules for Dingo, a software package with Python, command-line, and server interfaces. It points agents to the project's full AGENTS.md instructions and records rules for its evaluators, data, imports, and dependencies.

In plain words
What is it for?
Use it when modifying Dingo evaluators, data handling, dependencies, the command-line tool, or its server.
Why use it?
It keeps changes consistent with the project's existing design and prevents common breakages, such as incorrect evaluator results or unavailable optional libraries.

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/migoxlab/dingo/project-architecture
Clone the repo
git clone --depth 1 https://github.com/MigoXLab/dingo

Made for: Cursor.

Per session 367 This file is loaded in full into every session.
When invoked 367 The same file — it is already loaded in full.
Security scan A 0 findings. 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 $0.00367 $0.00367
Opus 5 $0.00183 $0.00183
Sonnet 5 $0.00073 $0.00073
Haiku 4.5 $0.00037 $0.00037

Measured 2d ago against content hash 7741506c699d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

project-architecture scanned grade A with 0 findings 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 2d 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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.cursor/rules/project-architecture.mdc · 21 lines

What it actually says

Dingo — Quick Reference

Full agent instructions: see AGENTS.md at project root (tech stack, directory structure, core concepts, development conventions, config maintenance rules).

Key Rules (always apply)

  1. Plugin architecture: All evaluators use @Model.rule_register() or @Model.llm_register() decorators. Never modify the registration system.
  2. Data model: Data uses extra = "allow" — do not add fixed fields unless absolutely necessary. Access optional fields with getattr(data, 'field', default).
  3. Lazy imports: Optional heavy dependencies (pyarrow, transformers, sqlalchemy, boto3, etc.) must use lazy imports inside methods, not top-level imports.
  4. Evaluator contract: Every evaluator eval() method must return EvalDetail with metric, status (bool), label (list), and reason (list).
  5. Dependencies: Core deps in requirements/runtime.txt, optional in requirements/datasource.txt. New optional deps → add to setup.py extras_require.
  6. MCP Server: Entry point is mcp_server.py, uses FastMCP + SSE. Environment variables for LLM config (OPENAI_API_KEY, etc.).
  7. Three interfaces: SDK (Python API), CLI (dingo eval), MCP Server. All share the same InputArgsExecutor pipeline.
  8. Testing: pytest test/scripts --ignore=test/scripts/data. Integration tests via CLI with configs in .github/env/.
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. 2d ago First seen · 21 lines · 367 tokens per session scan A 7741506c699d

Subscribe to this mod's changes

project-architecture is a cursor rule published in the GitHub repository MigoXLab/dingo (752 stars, last pushed 5d ago), licensed Apache-2.0. It adds 367 tokens to every session, about $0.0018 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.