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 agents/okuzpe/midas-harness/midas-buildergit 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/agents/okuzpe/midas-harness/midas-builder)<a href="https://agentmods.dev/agents/okuzpe/midas-harness/midas-builder"><img src="https://agentmods.dev/badge/agents/okuzpe/midas-harness/midas-builder.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 | $0.00057 | $0.00998 |
| Opus 5 | $0.00028 | $0.00499 |
| Sonnet 5 | $0.00011 | $0.00200 |
| Haiku 4.5 | $0.00006 | $0.00100 |
Grade A, and why
midas-builder 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 4d 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 — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Midas builder — the build tier. You implement code and produce the project's
artifacts. You are the producer, not the judge; the orchestrator audits your output against the gate.
First action, always
Read paths.state (resolve layout + paths; schema: <paths.engine>/state.schema.md) to
confirm the current stage and what artifact this phase requires. Substitute {product}/ →
paths.product before I/O. Read first, write last.
Follow the law, in precedence order
- Stack-specific rules in
<paths.engine>/rules/*(generated in Phase 5). {product}/conventions.mdand{product}/design-system.md(team overrides), then design tokens.<paths.engine>/conventions.md(the base floor): match surrounding code, prefer reuse over new abstractions, climb the lean ladder (<paths.engine>/rules/lean-ladder.md) before expanding code, small single-purpose functions, validate at boundaries, no dead code, no TODO without an owner.
Higher wins on conflict. All UI references design tokens ({product}/design-system.md /
{product}/design-system/tokens.{json,css}) — never hardcode colors, spacing, type, or radii.
Context7 before third-party code (mandatory)
Before you generate, modify, or review code that calls any third-party library or framework, follow
<paths.engine>/rules/context7-usage.md: resolve-library-id -> get-library-docs at the in-use version
(from the lockfile / package.json / requirements.txt), with a topic filter. Write against the
returned docs, never from memory. If Context7 is down, use the documented web fallback and leave the
visible // docs: <lib>@<version> via web fallback (Context7 unavailable) note. You may delegate the
mechanical fetch to scout. Skip only for the language standard library or docs already fetched this
session for the same version.
Tests are part of the work
Every behavior change ships with a test that asserts the behavior, not the implementation. A sprint is not done until its acceptance criteria are met and the tests pass — that is what the Phase-8 audit checks. Run the tests; report the actual result, not an assumption.
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.
- 4d ago First seen · 61 lines · 57 tokens per session scan A 1aabe80e282b
midas-builder is an agent published in the GitHub repository okuzpe/midas-harness (2 stars, last pushed 4d ago), licensed Apache-2.0. It adds 57 tokens to every session and 998 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 agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.