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 skills add Exponentially-Platform/Rapidly-MCP --skill test-an-ideagit clone --depth 1 https://github.com/Exponentially-Platform/Rapidly-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/skills/exponentially-platform/rapidly-mcp/test-an-idea)<a href="https://agentmods.dev/skills/exponentially-platform/rapidly-mcp/test-an-idea"><img src="https://agentmods.dev/badge/skills/exponentially-platform/rapidly-mcp/test-an-idea/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/exponentially-platform/rapidly-mcp/test-an-idea"><img src="https://agentmods.dev/badge/skills/exponentially-platform/rapidly-mcp/test-an-idea.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00123 | $0.00748 |
| Opus 5.5 | $0.00049 | $0.00299 |
| Sonnet 5.5 | $0.00025 | $0.00150 |
| Haiku 4.5 | $0.00012 | $0.00075 |
Grade A, and why
test-an-idea 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 13d 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 — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Test an idea before building it
Rapidly applies Pretotyping, Alberto Savoia's method for testing demand before you build. A pretotype fakes the essential part of the product and measures what real customers do, not what they say they'll do. Rapidly does the method work and saves it with the idea, so the team can come back to it.
Run it
- Check that the Rapidly tools are available, for example
test_ideaandfind_ideas. If they aren't, run thesetupskill first. - If the user has an idea, call
test_ideawith the idea in their words: who the customer is, the problem and the proposed change. Add any facts, sources and constraints they've already given you. - If the user has a company but no idea yet, call
find_ideaswith a short brief that includes the company name and website. Use the number of ideas they asked for, or one. Then calltest_ideafor each idea they want to test. - When
test_ideareturnsstatus: in_progress, call it again straight away with the sameproject_id. Keep going until it returnscompleteor a clearly explainedincompleteresult. Don't ask the user which stage to run or whether to save. Rapidly saves each stage itself. - Show the user the recommended first experiment, its pass mark, its go and no-go rule, and the build prompt exactly as Rapidly returned it. Mention that the other experiment options are saved with the idea.
What to tell the user
- The pass mark is a proposal to test, not evidence. Real customers provide the evidence.
- Next step: build the test asset from the build prompt, put it in front of real customers, then come back with what happened. The
record-a-test-resultskill takes it from there. - Generated facts and cited sources need a human check before anyone relies on them.
Don't expand the build prompt into a full product, dashboard or launch plan. The point is the smallest test that could change the decision.
The method in brief
- A Lean Canvas that names the customer, their problem and the proposed solution.
- The riskiest assumption written as a market engagement hypothesis: who will do what, if offered what.
- The same hypothesis made measurable as an XYZ hypothesis: at least X% of Y will do Z.
- The cheapest named technique that tests it, such as a Fake Door, Pinocchio, Mechanical Turk, Provincial, Infiltrator or Imposter test, with the pass mark and decision rule set before the test starts.
- A real test with real customers, then a decision to continue, change or stop.
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.
- 13d ago First seen · 37 lines · 123 tokens per session scan A 027e448f602f
test-an-idea is a skill published in the GitHub repository Exponentially-Platform/Rapidly-MCP (2 stars, last pushed 6d ago), licensed MIT. It adds 123 tokens to every session and 748 once invoked, about $0.0005 per session on Opus 5.5. 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 2026-10-10.
Other skills, from other repositories
featurevisor
Author, test, query and integrate Featurevisor feature flags, A/B experiments and remote configuration. Use in Featurevisor projects (featurevisor.config.js) for features, segments, attributes, global variables and compositions, conditions, rollouts, test specs and scenarios, datafiles, targets, Sets and promotions…
researcher
Use when user wants to optimize or tune something measurable through repeated experiments — "make X faster", "improve [metric]", "find best config", "iterate overnight", "run experiments until it hits N". Triggers on quantitative goals (build time, latency, pass rate, accuracy) or qualitative ones scoreable against a…
ab-testing
Use when designing or analyzing a controlled experiment — falsifiable hypothesis, sample size from an MDE, reading significance/CI/power, CUPED, or rescuing tests that won't go significant. NOT recurring metric tracking (that is analytics), NOT north-star/KPI trees (that is kpi-framework), NOT projecting metrics…
A/B Test Validation
Validating A/B test implementations including traffic splitting accuracy, statistical significance calculation, metric tracking, and experiment cleanup.
idea-os
Five-phase pipeline (triage → clarify → research → PRD → plan) that turns a raw idea into four linked files: clarifying questions, deep research, a PRD with non-goals and metrics, and a phased execution plan with mermaid user journey and kill criteria.
idea-check
Pre-build reality check. Scans GitHub, HN, npm, PyPI, and Stack Overflow for existing competitors before you build. Use when starting a new project, evaluating a side project idea, or doing a build-vs-buy decision.