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 Rheinmir/setup --skill dym-harness-sdk-quickstartgit clone --depth 1 https://github.com/Rheinmir/setupWrote 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/rheinmir/setup/dym-harness-sdk-quickstart)<a href="https://agentmods.dev/skills/rheinmir/setup/dym-harness-sdk-quickstart"><img src="https://agentmods.dev/badge/skills/rheinmir/setup/dym-harness-sdk-quickstart/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/rheinmir/setup/dym-harness-sdk-quickstart"><img src="https://agentmods.dev/badge/skills/rheinmir/setup/dym-harness-sdk-quickstart.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 | $0.00053 | $0.00832 |
| Opus 5.5 | $0.00021 | $0.00333 |
| Sonnet 5.5 | $0.00011 | $0.00166 |
| Haiku 4.5 | $0.00005 | $0.00083 |
Grade A, and why
dym-harness-sdk-quickstart 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 yesterday.
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 — 38 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill: dym-harness-sdk-quickstart — dựng agent lắp sẵn
When to use
- Muốn một agent "đủ đồ" (shell, file, web, subagent, todo, memory, session, offload context) trong 1 dòng, không tự viết agent loop.
- Muốn xem harness lắp gì TRƯỚC khi tốn tiền gọi model.
- Sinh ra: một
strands.AgentTHƯỜNG (không phải lớp bọc riêng) — mọi thứ SDK làm được vẫn làm được sau khi dựng.
Steps
- Venv cô lập, Python ≥ 3.10 (Python 3.9 hệ thống của macOS không cài được):
python3.13 -m venv .venv-strands && .venv-strands/bin/pip install -q strands-harness # provider khác Bedrock: strands-harness[anthropic] | [openai] | [gemini] | [ollama] | [litellm] - Dựng agent KHÔNG gọi model (đo 2026-09-28, rc 0):
from strands_harness import create_harness a = create_harness(session=False, memory=False) print(sorted(a.tool_names)) # ['edit','programmatic_tool_caller','read','retrieve_context','retrieve_offloaded_content', # 'shell','strands_manage_background_task','subagent','todo_write','web_fetch','write'] - Chạy thật (tốn tiền):
a("Find the slowest test in this repo"). Mặc định model = Bedrock Opus 5 → cần AWS credential; đổi bằngmodel="anthropic/claude-fable-5"(chuỗiprovider/name). - Tham số chính (
create_harness, keyword-only):model·effort(auto|off|minimal..max) ·instructions(ghép SAU contract của harness) ·tools·mcp_servers·builtin_tools·session(mặc định./.agent/sessions) ·memory(mặc định./.agent/memory) ·skills·builtin_plugins·interventions·**agent_kwargs(giá trị SDK tường minh luôn THẮNG giá trị mặc định của harness). - Bật trace: đặt
OTEL_TRACES_EXPORTER=otlp(hoặcconsole) trước khi dựng; không đặt thì telemetry tắt.
Rules
- Bẫy đã đo: model không có web search native (Bedrock) → harness chỉ in cảnh báo rồi bỏ tool
web_search, không lỗi. Cần search thìbuiltin_tools={'web_search': 'exa'}. - Bẫy đã đo:
session/memorymặc định BẬT và ghi vào./.agent/của cwd — thử nghiệm thì tắt (session=False, memory=False) để không bẩn repo. - Tool trùng tên giữa 5 nguồn (consumer · builtin · MCP · plugin · subagent) → raise ngay lúc dựng; riêng tool MCP chỉ biết tên sau khi kết nối nên lọt qua bước kiểm này.
- Memory extract chạy nền: script chạy ngắn phải gọi shutdown, nếu không có thể mất dữ liệu.
- Thứ cho production: bản harness đang 0.x và bật mặc định
ContextManagercủastrands.experimental— không cam kết semver.
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.
- yesterday First seen · 38 lines · 53 tokens per session scan A f16da4e6ba28
dym-harness-sdk-quickstart is a skill published in the GitHub repository Rheinmir/setup (11 stars, last pushed 20d ago), licensed MIT. It adds 53 tokens to every session and 832 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 2026-11-07.
dym-harness-sdk-quickstart compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| dym-harness-sdk-quickstartRheinmir/setup | Skill | 11 | 20d ago | MIT | A | 53 | Claude Code |
| matlabK-Dense-AI/scientific-agent-skills | Skill | 56k | 6d ago | MIT | A | 44 | Claude Code |
| pennylaneK-Dense-AI/scientific-agent-skills | Skill | 56k | 6d ago | MIT | A | 74 | Claude Code |
| holoscan-install-wheelNVIDIA/skills | Skill | 3.9k | yesterday | Apache-2.0 | A | 34 | Claude Code, Codex |
| future-self-liteTHU-MAIC/OpenMAIC | Skill | 43k | yesterday | MIT | A | 20 | Claude Code, Codex |
| rocm-kernelshuggingface/kernels | Skill | 783 | yesterday | Apache-2.0 | A | 93 | Claude Code |
| typing-exclusion-workergetsentry/skills | Skill | 1.1k | 2d ago | Apache-2.0 | A | 57 | Claude Code |