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/anymouschina/tapcanvas/hardnessgit clone --depth 1 https://github.com/anymouschina/TapCanvasWrote 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/anymouschina/tapcanvas/hardness)<a href="https://agentmods.dev/agents/anymouschina/tapcanvas/hardness"><img src="https://agentmods.dev/badge/agents/anymouschina/tapcanvas/hardness.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 | $0.00000 | $0.09089 |
| Opus 5 | $0.00000 | $0.04544 |
| Sonnet 5 | $0.00000 | $0.01818 |
| Haiku 4.5 | $0.00000 | $0.00909 |
Grade A, and why
hardness 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 — 382 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skip to main contentSkip to footer
* Research
- Economic Futures
- Commitments
- Learn
- News
Harness design for long-running application development
Published Mar 24, 2026
Harness design is key to performance at the frontier of agentic coding. Here's how we pushed Claude further in frontend design and long-running autonomous software engineering.
Written by Prithvi Rajasekaran, a member of our Labs team.
Over the past several months I’ve been working on two interconnected problems: getting Claude to produce high-quality frontend designs, and getting it to build complete applications without human intervention. This work originated with earlier efforts on our frontend design skill and long-running coding agent harness, where my colleagues and I were able to improve Claude’s performance well above baseline through prompt engineering and harness design—but both eventually hit ceilings.
To break through, I sought out novel AI engineering approaches that held across two quite different domains, one defined by subjective taste, the other by verifiable correctness and usability. Taking inspiration from Generative Adversarial Networks (GANs), I designed a multi-agent structure with a generator and evaluator agent. Building an evaluator that graded outputs reliably—and with taste—meant first developing a set of criteria that could turn subjective judgments like “is this design good?” into concrete, gradable terms.
I then applied these techniques to long-running autonomous coding, carrying over two lessons from our earlier harness work: decomposing the build into tractable chunks, and using structured artifacts to hand off context between sessions. The final result was a three-agent architecture—planner, generator, and evaluator—that produced rich full-stack applications over multi-hour autonomous coding sessions.
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 · 382 lines · 0 tokens per session scan A 6a9dc20b5f51
hardness is an agent published in the GitHub repository anymouschina/TapCanvas (590 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 9,089 tokens. 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 2026-08-30.
Other agents, from other repositories
d3-expert
Expert in D3.js for creating custom, interactive data visualizations with SVG, Canvas, and HTML. Specializes in D3 v7+ with ES modules, selections, data binding, scales, transitions, force simulations, hierarchical layouts, geographic projections, and performance optimization for large datasets. Use for any data…
lattice-architect
The visionary. Designs the public API of a Lattice package before a line of it is implemented, and says what the kit is missing. Use before any builder starts, and whenever two packages disagree about a shape.
lattice-builder
The developer. Implements exactly one Lattice package to its RFC — source, tests, benchmarks and README — and leaves the repo green. Owns only the paths its task names.
lattice-auditor
The adversary. Audits finished Lattice packages against the constitution and hunts for the bug the tests were written to miss. Reports, and fixes only what its task authorises.
codex
Agent "codex" from free-revalution/AIGC-Detector-Pro, covering codex cli entry template, location, format, content and notes.
Demonstrate
Agent for demonstrating VS Code features.