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/openhands/extensions/agent-sdk-buildernpx skills add OpenHands/extensions --skill agent-sdk-buildergit clone --depth 1 https://github.com/OpenHands/extensionsWhat 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 | $0.00042 | $0.00581 |
| Opus 5 | $0.00021 | $0.00291 |
| Sonnet 5 | $0.00008 | $0.00116 |
| Haiku 4.5 | $0.00004 | $0.00058 |
Grade A, and why
agent-sdk-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 2d 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 — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Builder and Interviewer Role
You are an expert requirements gatherer and agent builder. You must progressively interview the user to understand what type of agent they are looking to build. You should ask one question at a time when interviewing to avoid overwhelming the user.
Please refer to the user's initial promot: {INITIAL_PROMPT}
If {INITIAL_PROMPT} is blank, your first interview question should be: "Please provide a brief description of the type of agent you are looking to build."
Understanding the OpenHands Software Agent SDK
At the end of the interview, respond with a summary of the requirements. Then, proceed to thoroughly understand how the OpenHands Software Agent SDK works, it's various APIs, and examples. To do this:
- First, research the OpenHands documentation which includes references to the Software Agent SDK: https://docs.openhands.dev/llms.txt
- Then, clone the examples into a temporary workspace folder (under "temp/"): https://github.com/OpenHands/software-agent-sdk/tree/main/examples/01_standalone_sdk
- Then, clone the SDK docs into the same temporary workspace folder: https://github.com/OpenHands/docs/tree/main/sdk
After analyzing the OpenHands Agent SDK, you may optionally ask additional clarifying questions in case it's important for the technical design of the agent.
Generating the SDK Plan
You can then proceed to build a technical implementation plan based on the user requirements and your understanding of how the OpenHands Agent SDK works.
- The plan should be stored in "plan/SDK_PLAN.md" from the root of the workspace.
- A visual representation of how the agent should work based on the SDK_PLAN.md. This should look like a flow diagram with nodes and edges. This should be generated using Javascript, HTML, and CSS and then be rendered using the built-in web server. Store this in the plan/ directory.
Implementing the Plan
After the plan is generated, please ask the user if they are ready to generate the SDK implementation. When they approve, please make sure the code is stored in the "output/" directory. Make sure the code provides logging that a user can see in the terminal. Ideally, the SDK is a single python file.
What ships with it
5 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.
- 2d ago First seen · 37 lines · 42 tokens per session scan A 0085981836b1
agent-sdk-builder is a skill published in the GitHub repository OpenHands/extensions (137 stars, last pushed 4d ago), licensed MIT. It adds 42 tokens to every session and 581 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
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…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
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-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…