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 hamzaPixl/pixl-ai --skill harnessgit clone --depth 1 https://github.com/hamzaPixl/pixl-aiWrote 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/hamzapixl/pixl-ai/harness)<a href="https://agentmods.dev/skills/hamzapixl/pixl-ai/harness"><img src="https://agentmods.dev/badge/skills/hamzapixl/pixl-ai/harness/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/hamzapixl/pixl-ai/harness"><img src="https://agentmods.dev/badge/skills/hamzapixl/pixl-ai/harness.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 | $0.00062 | $0.01753 |
| Opus 5.5 | $0.00025 | $0.00701 |
| Sonnet 5.5 | $0.00012 | $0.00351 |
| Haiku 4.5 | $0.00006 | $0.00175 |
Grade A, and why
harness 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 1mo 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 — 151 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Harness
GAN-inspired generate-evaluate-iterate loop. A generator agent builds the application, then an independent evaluator agent grades the output on four axes. If any score falls below threshold, the evaluator's critique feeds back into the generator for another pass. The loop continues until all scores pass or max iterations are reached.
Required References
Before starting, read ALL of these:
references/grading-rubric.md-- 4-axis scoring criteria (design quality, originality, craft, functionality)references/anti-rationalization.md-- rules for honest evaluation, common rationalization trapsreferences/integrity-protocol.md-- no-hack rules, zero-assumption debugging, consensus validation, escalation protocol
Step 1: Input & Configuration
Accept a 1-4 sentence product description as the skill argument.
Ask the user (via AskUserQuestion) for optional configuration overrides. Use these defaults if the user skips:
| Parameter | Default | Description |
|---|---|---|
| Threshold | 7/10 | Minimum score on each axis to pass |
| Max iterations | 5 | Maximum generate-evaluate cycles before hard stop |
| Budget cap | $150 | Total spend limit across all iterations |
| Aesthetic direction | (none) | Optional style guidance (e.g., "museum quality", "minimal dark mode") |
Step 2: Execute via Engine Workflow (preferred path)
If the pixl CLI is available, delegate to the engine DAG:
pixl workflow run harness --prompt "<user prompt>" --yes
This runs the full DAG: plan -> generate -> evaluate -> score-gate -> (loop on failure).
Use --stream for real-time progress output. Monitor cost with pixl cost summary after completion.
If the engine workflow succeeds, skip to Step 4 (report).
Step 3: Fallback -- Direct Agent Loop (when pixl CLI unavailable)
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago First seen · 151 lines · 62 tokens per session scan A 310674b4fe14
harness is a skill published in the GitHub repository hamzaPixl/pixl-ai (2 stars, last pushed 6mo ago), licensed MIT. It adds 62 tokens to every session and 1,753 once invoked, about $0.0002 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 2026-09-03.
Other skills, from other repositories
verify-omnigent
Drive Omnigent the way a user does and prove a behavior with recorded evidence, using an isolated server, runner, and mock model plus a feature map of every user entry point. Load before reproducing a user-facing bug, before claiming a fix works, or when reviewing whether a change covered every surface (session vs.…
test-master
Generates test files, creates mocking strategies, analyzes code coverage, designs test architectures, and produces test plans and defect reports across functional, performance, and security testing disciplines. Use when writing unit tests, integration tests, or E2E tests; creating test strategies or automation…
eval-dataset-design
A guide to creating reliable test sets for AI agents. Each evaluation task includes the request, the simulated user's behaviour, a reset starting state, and a success check that can be independently verified.
dogfood
Systematically explore and test a mobile app on iOS/Android with agent-device to find bugs, UX issues, and other problems. Use when asked to dogfood, QA, exploratory test, find issues, bug hunt, or test this app on mobile.
cua-driver
Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.
critique
Adversarial design critique of the current work. Render it, look at it, and argue for rejection. Run after the gates are green, never instead of them.