midas-harness: Skill for Claude Code

.agents/skills/market-research/SKILL.md

market-research is a skill for Claude Code, Codex from okuzpe/midas-harness. It costs 81 tokens per session (1,430 once invoked), scanned A, original, Apache-2.0.

A market-validation workflow for checking whether a clarified product idea addresses a real problem for a real audience. It uses cited web research to study competitors and market risks.

In plain words
What is it for?
Use it to define research questions, gather and verify online evidence, build a competitor comparison, identify differences and risks, and save the findings in a market.md file.
Why use it?
It reduces the risk of building on unsupported assumptions or missing important competitors. It also makes claims traceable to their sources.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions subagents; installed under .agents/ (shared by several agents); mentions AGENTS.md.

This is okuzpe/midas-harness's own configuration. It tells Claude Code and Codex how to work on midas-harness itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything midas-harness configures →

Reuse

Borrowing it

Nothing to install: this file belongs to okuzpe/midas-harness. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/okuzpe/midas-harness/main/.agents/skills/market-research/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/okuzpe/midas-harness

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for market-research

README.md
[![agentmods](https://agentmods.dev/badge/skills/okuzpe/midas-harness/market-research/github.svg)](https://agentmods.dev/skills/okuzpe/midas-harness/market-research)
Your own site
<a href="https://agentmods.dev/skills/okuzpe/midas-harness/market-research"><img src="https://agentmods.dev/badge/skills/okuzpe/midas-harness/market-research/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.

agentmods 80×15 button for market-research

Your own site · 80×15
<a href="https://agentmods.dev/skills/okuzpe/midas-harness/market-research"><img src="https://agentmods.dev/badge/skills/okuzpe/midas-harness/market-research.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 81 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,430 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00081 $0.01430
Opus 5 $0.00041 $0.00715
Sonnet 5 $0.00016 $0.00286
Haiku 4.5 $0.00008 $0.00143

Measured 6d ago against content hash 0fac8d37aa77, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

market-research 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 6d 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.

.agents/skills/market-research/SKILL.md · 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.

market-research — Phase 2

Guard + state: <paths.engine>/templates/skill-state-ritual.md (+ AGENTS.md § Safety / Path resolution). Precondition: contextualize passed (user/problem/metric/non-goals clear). Blocking opens → /contextualize. Playbook: <paths.engine>/pipeline/2-market-research.md

Validate that the clarified idea addresses a real problem with a real audience, and map the competitive landscape with citations. The producer gathers and synthesizes; the orchestrator frames the questions and audits the gate — it does not rubber-stamp its own report.

Does / Does not

Does Does not
Desk research with cited claims → {product}/market.md Fabricate citations or leave material claims uncited
Competitor matrix + differentiation + top 3 risks + demand verdict Field interviews as a hard wall (that is Phase 3)
Stop if Phase-1 blockers remain Rubber-stamp the report as the gate auditor

When NOT

  • track: liteSTOP. Market research is skipped on Lite. Point at /midas-status (never run Phase 2 as Next; do not recover via /market-research). See <paths.engine>/pipeline/lite.md.
  • Blocking open questions remain → /contextualize.
  • User wants go/no-go / monetization → /business-plan after this gate (track: full).
  • “No time to research” → still required on full track; use scout fan-out + strike uncited claims rather than inventing.

Anti-rationalization: a competitor list without a demand verdict (strong/mixed/weak + evidence) does not pass the exit gate.

Steps

  1. Derive research questions from {product}/idea.md: market-size signals, direct competitors, substitutes/alternatives, pricing norms, distribution channels, regulatory/compliance constraints, and demand signals — evidence that the problem is real and people pay to solve it (competitor traction/reviews/funding, complaints in forums/Reddit/app-store reviews, search/community interest, what people already pay for substitutes). This is the part you CAN validate from the desk.
  2. Fan out the research. Dispatch scout subagents (Haiku) to WebSearch + fetch each question; for any technology/landscape facts, use Context7. If an external deep-research skill is installed in the host tool, you may delegate the fan-out to it — it is not part of the Midas engine.
  3. Adversarially verify. Every material claim must cite a source URL. Strike or flag uncited claims. Distinguish primary sources from blog hearsay.
  4. Synthesize (build tier) into a competitor matrix (who, what, price, gap), a one-paragraph differentiation thesis, the top 3 market risks with their early signals, and a frank demand verdictstrong / mixed / weak desk-signal — citing the evidence behind it (traction, pain complaints, search interest, willingness-to-pay). State plainly what the desk can and cannot prove: it shows a market exists, not that these customers will pay — that is field validation (Phase 3).
  5. Write {product}/market.md from <paths.engine>/templates/market.md — keep those headings (## Market overview, ## Target segment, ## Competitive landscape, ## Differentiation thesis, ## Demand signals, ## Top 3 risks, ## Sources). Do not invent a parallel outline. Update paths.state (read-modify-write) (market_research: in_progress → leave the gate verdict to the orchestrator).

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. 6d ago Changed · +5 lines 0fac8d37aa77
  2. 11d ago First seen · 84 lines · 81 tokens per session scan A 58758555df34

Subscribe to this mod's changes

market-research is a skill published in the GitHub repository okuzpe/midas-harness (2 stars, last pushed 6d ago), licensed Apache-2.0. It adds 81 tokens to every session and 1,430 once invoked, about $0.0004 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.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

insight-error-page

Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…

vercel/next.js · 83 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens