PenguinHarness is a local-first platform in which multiple AI agents create, evaluate, optimize, and deploy agent applications. It is for people building AI software who want agents to generate applications and improve their own behavior through skills.
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 Prism-Shadow/penguin-harness --skill ape-rsi-toolkitgit clone --depth 1 https://github.com/Prism-Shadow/penguin-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/prism-shadow/penguin-harness/ape-rsi-toolkit)<a href="https://agentmods.dev/skills/prism-shadow/penguin-harness/ape-rsi-toolkit"><img src="https://agentmods.dev/badge/skills/prism-shadow/penguin-harness/ape-rsi-toolkit/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/prism-shadow/penguin-harness/ape-rsi-toolkit"><img src="https://agentmods.dev/badge/skills/prism-shadow/penguin-harness/ape-rsi-toolkit.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 | $0.00000 | $0.03405 |
| Opus 5.5 | $0.00000 | $0.01362 |
| Sonnet 5.5 | $0.00000 | $0.00681 |
| Haiku 4.5 | $0.00000 | $0.00341 |
Grade A, and why
ape-rsi-toolkit 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 — 158 lines — stays where its author put it; the contents beside it link to each section on GitHub.
APE
Large Language Models are Human-Level Prompt Engineers (Zhou et al., University of Toronto, Vector Institute and University of Waterloo, 2022; paper v2, §3 and §5.3), code at keirp/automatic_prompt_engineer. Verified implementation revision eac521c79a78965245ce7745dcc9f6b0792c7ec7: automatic_prompt_engineer/ape.py, generate.py, evaluate.py and template.py, experiments/run_instruction_induction.py and experiments/evaluation/instruction_induction/exec_accuracy.py. This Skill runs non-iterative forward-generation APE with execution scoring, the paper's default: a proposer infers candidate instructions from input/output demonstrations in one pass, every candidate is measured on the selection cases, and the best is kept. It implements neither the optional iterative Monte Carlo resampling of §5.3 nor bandit (UCB) allocation nor log-likelihood scoring; the public simple_ape helper uses UCB, so never claim that helper's profile. Run the method the way the paper does; every departure is a declared adaptation reported at the end.
Before you start
This Skill is the experiment's controller and needs run_subagent; if that tool is missing you are a subagent: stop and say the Skill needs a top-level Session. If the request only names this Skill, ask for the Target, the Benchmark and the demonstrations, and offer the demo task. Otherwise resolve the inputs below, asking only for one the method needs and the request does not give:
test_agent_id: the Target, an existing Agent of this Project, orcreate: <id>for a fresh one. Never the Agent running this Skill: its workers are this Agent.benchmark_id: a published Benchmark of this Project; its cases are APE's selection set (the source'seval_data). When the user asks for the demo task, build it fromreferences/demo-task.mdfirst: it fixes the Target, the Benchmark, the demonstrations and the budget, so ask nothing.case_ids: every case unless the user names a subset.runs: Runs per Case, 1 unless given.- Demonstrations: the input/output pairs the proposer induces from (the source's
prompt_gen_data).- The demo task ships five.
- Otherwise ask for a file of pairs, or for explicit permission naming the Cases whose gold answers may serve as outputs; those Cases then leave the selection set. A private answer is not authorized because it exists.
- Never stand failed Target outputs or scores in for demonstrations: that is not APE. Until you have them, wait: open no Rubric and generate nothing.
- Keep demonstrations and selection cases apart, declare any overlap, and never use test cases for either.
- The Target's runtime, fixed as
references/evaluation.mdsays. The proposer is this Session's model. - The method's parameters, with the values of
run_instruction_induction.py:
What ships with it
3 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.
- 4d ago First seen · 158 lines · 0 tokens per session scan A 1e612921d0d9
ape-rsi-toolkit is a skill published in the GitHub repository Prism-Shadow/penguin-harness (2,753 stars, last pushed today), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 3,405 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 2026-10-13.
Other skills, from other repositories
deepseek-harness
Use when building AI agent applications with a plugin-based architecture — Web UI, CLI, Python SDK, Cordis plugin system, multi-model orchestration. DeepSeek Harness (dsh): open-source agent harness by DeepSeek AI where everything is a plugin, powered by Cordis for spatiotemporal composability.
LLM
Implement large language model (LLM) chat completions using the z-ai-web-dev-sdk. Use this skill when the user needs to build conversational AI applications, chatbots, AI assistants, or any text generation features. Supports multi-turn conversations, system prompts, and context management.
ASR
Implement speech-to-text (ASR/automatic speech recognition) capabilities using the z-ai-web-dev-sdk. Use this skill when the user needs to transcribe audio files, convert speech to text, build voice input features, or process audio recordings. Supports base64 encoded audio files and returns accurate text…
VLM
Implement vision-based AI chat capabilities using the z-ai-web-dev-sdk. Use this skill when the user needs to analyze images, describe visual content, or create applications that combine image understanding with conversational AI. Supports image URLs and base64 encoded images for multimodal interactions.
deepseek-harness
Use this skill whenever the user wants to call DeepSeek V4-Pro / V4-Flash (or its legacy aliases deepseek-chat / deepseek-reasoner), or you see code that imports from openai import OpenAI with baseurl="https://api.deepseek.com". This skill teaches you the 10 protocol contract rules required to avoid the 16 documented…
yait-aichain
Build AI features in Python — an LLM call, a multi-step pipeline, the same work over many items, or an agent that uses tools (web search, page fetching, your own functions, any REST API or MCP server, a working folder), asks a human before risky actions, stays within a money budget, or pauses and resumes — on any…