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 skills/okuzpe/midas-harness/contextualizenpx skills add okuzpe/midas-harness --skill contextualizegit clone --depth 1 https://github.com/okuzpe/midas-harnessWrote 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/okuzpe/midas-harness/contextualize)<a href="https://agentmods.dev/skills/okuzpe/midas-harness/contextualize"><img src="https://agentmods.dev/badge/skills/okuzpe/midas-harness/contextualize.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 5.1 | $0.00070 | $0.01384 |
| Opus 5 | $0.00035 | $0.00692 |
| Sonnet 5 | $0.00014 | $0.00277 |
| Haiku 4.5 | $0.00007 | $0.00138 |
Grade A, and why
contextualize 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 today.
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 — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.
contextualize — Phase 1: the gap loop
Guard + state:
<paths.engine>/templates/skill-state-ritual.md(+AGENTS.md§ Safety / Path resolution). Prompt tool:AskQuestion. On Claude Code, fall back toAskUserQuestionif AskQuestion is not wired.
This is Midas's signature phase. A raw idea is full of unstated assumptions; building on them is the
most expensive mistake. Phase 1 systematically surfaces every blocking unknown and resolves it
with the user before any market, business, or architecture work. Playbook:
<paths.engine>/pipeline/1-contextualize.md.
Precondition: paths.state at stage: contextualize, with {product}/idea.md from Phase 0.
Does / Does not
| Does | Does not |
|---|---|
| Surface and resolve blocking unknowns with the user | Invent answers or silently assume blockers away |
Fold answers into {product}/idea.md + {product}/open-questions.md |
Overwrite the Phase-0 verbatim raw-idea block |
| Loop until zero blockers (or deferred + logged assumption) | Dump >4 questions per batch; start market/arch early |
When NOT
- Phase 0 incomplete (no verbatim idea / pitch / mode) →
/idea-intake. - Zero blocking opens already and gate passed → next is
/market-research(track: fullonly). Whentrack: lite, next is/midas-status(overlay → remaining Idea+Plan or/plan-sprints) — never/market-research. - User wants status only →
/midas-status.
Anti-rationalization: “we’ll figure it out in build” is not a deferred assumption — deferrals
need an explicit user-accepted written assumption in {product}/open-questions.md.
The loop
Repeat until there are zero blocking open questions:
- Generate questions. From
{product}/idea.md, enumerate the unknowns that, if answered wrong, would invalidate downstream phases. Cover at least: target user, the problem (and evidence it's real), the success metric (one measurable signal of "this worked"), and non-goals (what the MVP deliberately won't do). - Rank by severity. Tag each
blocking(a later phase cannot proceed correctly without it) ornon-blocking(nice to know; defer). Onlyblockingquestions gate this phase. - Ask in batches. Use
AskQuestionto ask the highest-severity blockers in small batches (≈3-4 at a time), each with a crisp default/option set so the user can answer fast. Do not dump 20 questions at once. - Fold answers in. Update
{product}/idea.md(v2) under## Contextualizedwith the resolved facts — sharpening the user, problem, metric, and non-goals sections. Keep the verbatim raw-idea block from Phase 0 intact. - Maintain the ledger. Keep
{product}/open-questions.mdcurrent from<paths.engine>/templates/open-questions.md: headings### BLOCKING/### DEFERRED/### RESOLVED. A deferred blocker is only allowed if the user explicitly accepts a written assumption in its place — that assumption is logged here and surfaced by/midas-status. - Re-rank and loop. New answers often spawn new questions. Regenerate, re-rank, and continue until
no
blockingquestion remainsopen.
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.
- today Changed · +1 lines f4f90c4d4dbb
- 5d ago First seen · 89 lines · 70 tokens per session scan A d0b3bd69d1bf
contextualize is a skill published in the GitHub repository okuzpe/midas-harness (2 stars, last pushed yesterday), licensed Apache-2.0. It adds 70 tokens to every session and 1,384 once invoked, about $0.0003 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 skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
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…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
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…