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 agentmods add rules/etiennebbeaulac/memory-mcp/design-test-suitegit clone --depth 1 https://github.com/EtienneBBeaulac/memory-mcpWrote 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/rules/etiennebbeaulac/memory-mcp/design-test-suite)<a href="https://agentmods.dev/rules/etiennebbeaulac/memory-mcp/design-test-suite"><img src="https://agentmods.dev/badge/rules/etiennebbeaulac/memory-mcp/design-test-suite.svg" alt="Measured on agentmods" 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 | $0.00018 | $0.01406 |
| Opus 5 | $0.00009 | $0.00703 |
| Sonnet 5 | $0.00004 | $0.00281 |
| Haiku 4.5 | $0.00002 | $0.00141 |
Grade A, and why
design test suite 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 5d 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.
How it starts
The opening of the file, as written. The whole thing — 137 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Design a structured manual test plan for the tool, MCP, feature, or capability the user describes. If no description was provided, ask for one before proceeding.
The plan must be saved as a markdown file — ask the user where to save it if not obvious from context.
Two modes
Choose based on what the user is testing:
- Capability test — verifies the tool works correctly across a range of inputs. Uses mechanical suites (exact steps) and behavioral scenarios (open-ended). The CONTINUOUS_TEST_PLAN.md in this repo is an example.
- Hypothesis validation — tests a specific assumption about agent behavior (e.g., "do agents act on warnings?", "does the tool description change phrasing?"). Uses isolated, single-hypothesis suites with pure transcript evaluation. The TIER1_VALIDATION_PLAN.md in this repo is an example.
Critical design rules
These apply to both modes and are derived from observed test failures.
Never embed debrief questions in Part 1
Agents read the full instructions before executing. Any question like "did you notice X?" or "did you phrase it as Y?" tells the agent what to look for before they start — which contaminates the exact behavior you're measuring.
In capability tests: deliver debrief questions as separate evaluator prompts, pasted to the agent after each suite or scenario completes.
In hypothesis validation tests: use no debrief at all in Part 1. Ask at most one targeted question post-suite, after cleanup. The question must not name the axis being measured (e.g., ask "what influenced how you phrased it?" not "did you use present tense?").
Use a single session with contamination controls, not separate sessions per scenario
Fresh sessions eliminate prior-context contamination but also eliminate useful learning contamination. A single session with ordering controls and per-scenario cleanup is strictly better:
- Put schema/mechanical tests first — agents learn the correct API shape before behavioral tests run, eliminating first-attempt schema failures
- Put warning-mechanics tests after behavioral tests — agents won't have seen warning patterns before making natural judgment calls
- Use cleanup commands between scenarios to prevent stored entries from cross-contaminating retrieval results
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.
- 5d ago First seen · 137 lines · 18 tokens per session scan A 4e82d053d60a
design test suite is a cursor rule published in the GitHub repository EtienneBBeaulac/memory-mcp (2 stars, last pushed 3mo ago), licensed MIT. It adds 18 tokens to every session and 1,406 once invoked, about $0.0001 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 2026-08-31.
Other cursor rules, from other repositories
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
cli-error-handling
CLI command error handling patterns.
prefer-direct-imports-over-module-mocks
Prefer extracting a testable core over vi.mock / vi.resetModules when unit tests need to reach production logic entangled with config, env, or singletons.
control-plane-descriptors
Control plane descriptor and instance implementation patterns.