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 almond-bot/harness-studio --skill skillgit clone --depth 1 https://github.com/almond-bot/harness-studioWrote 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/almond-bot/harness-studio/skill)<a href="https://agentmods.dev/skills/almond-bot/harness-studio/skill"><img src="https://agentmods.dev/badge/skills/almond-bot/harness-studio/skill/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/almond-bot/harness-studio/skill"><img src="https://agentmods.dev/badge/skills/almond-bot/harness-studio/skill.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 | $0.00084 | $0.01901 |
| Opus 5.5 | $0.00034 | $0.00760 |
| Sonnet 5.5 | $0.00017 | $0.00380 |
| Haiku 4.5 | $0.00008 | $0.00190 |
Grade A, and why
almond-harness-studio 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 — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.
almond-harness-studio
Author a *.harness.json file, validate it, fetch its parts from the distributor, and export a vector-PDF drawing sheet the user can send to a harness manufacturer. Never draw harnesses by hand or with other tools; use this workflow.
Components are always real, orderable parts sourced from LCSC, Mouser, or Digi-Key — there are no free-form components (wire and bulk coverings are the only exceptions). Every connector, terminal, and accessory carries a part reference like { "vendor": "lcsc", "number": "C30170181" }.
Workflow
- Write the harness as
<name>.harness.json(format below). Look up real distributor part numbers for every connector/terminal/accessory — search the vendor sites if the user didn't give you one, and confirm your picks with the user when uncertain. - Validate and fix until clean:
npx almond-harness-studio validate ./my-harness.harness.json
- Fetch distributor data (description, MPN, photo, price, stock) into the file:
npx almond-harness-studio parts fetch ./my-harness.harness.json
LCSC needs no API key. Mouser and Digi-Key keys come from the user:
npx almond-harness-studio config set mouser.apiKey <key>
npx almond-harness-studio config set digikey.clientId <id>
npx almond-harness-studio config set digikey.clientSecret <secret>
(or env vars MOUSER_API_KEY, DIGIKEY_CLIENT_ID, DIGIKEY_CLIENT_SECRET)
- Export the drawing — resolved parts render with product photos and a sourced BOM:
npx almond-harness-studio export ./my-harness.harness.json -o ./my-harness.pdf
- If the manufacturer wants spreadsheets, export the wiring table and BOM as CSV:
npx almond-harness-studio tables ./my-harness.harness.json
- For a live preview the user can watch while you iterate:
npx almond-harness-studio dev ./harnesses-folder
Validation errors include the JSON path and an actionable message. Unresolved parts are reported as warnings until parts fetch runs. Do not hand the user a PDF until validation passes with no errors and all parts are fetched.
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.
- 2d ago Changed 675676173e9e
- 3d ago Changed 306f4ed29ead
- 1mo ago Changed 13d79814b6e1
- 2mo ago First seen · 112 lines · 84 tokens per session scan A 5654c5fc2e8e
almond-harness-studio is a skill published in the GitHub repository almond-bot/harness-studio (6 stars, last pushed yesterday), licensed MIT. It adds 84 tokens to every session and 1,901 once invoked, about $0.0003 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 2026-08-31.
almond-harness-studio compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| almond-harness-studioalmond-bot/harness-studio | Skill | 6 | yesterday | MIT | A | 84 | Claude Code, Codex |
| solidworks-vibecadwzyn20051216/solidworks-automation-skill | Skill | 1.8k | 2d ago | MIT | A | 96 | Claude Code, Codex |
| bdf_font_converterespressif/esp-claw | Skill | 2.4k | 5d ago | Apache-2.0 | A | 27 | Claude Code, Codex |
| fluidcad-assemblyFluid-CAD/FluidCAD | Skill | 693 | yesterday | MIT | A | 154 | Claude Code, Codex |
| gen-assemblypartcad/partcad | Skill | 515 | yesterday | Apache-2.0 | A | 111 | Claude Code |
| gen-partpartcad/partcad | Skill | 515 | yesterday | Apache-2.0 | A | 114 | Claude Code |
| gen-sketchpartcad/partcad | Skill | 515 | yesterday | Apache-2.0 | A | 201 | Claude Code |