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 Sounio-lang/sounio --skill automatic-differentiation-advancedgit clone --depth 1 https://github.com/Sounio-lang/sounioWrote 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/sounio-lang/sounio/automatic-differentiation-advanced)<a href="https://agentmods.dev/skills/sounio-lang/sounio/automatic-differentiation-advanced"><img src="https://agentmods.dev/badge/skills/sounio-lang/sounio/automatic-differentiation-advanced/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/sounio-lang/sounio/automatic-differentiation-advanced"><img src="https://agentmods.dev/badge/skills/sounio-lang/sounio/automatic-differentiation-advanced.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00050 | $0.01181 |
| Opus 5 | $0.00025 | $0.00590 |
| Sonnet 5 | $0.00010 | $0.00236 |
| Haiku 4.5 | $0.00005 | $0.00118 |
Grade A, and why
automatic-differentiation-advanced 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 9d 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 — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Advanced Automatic Differentiation
When to use this skill
Use this skill when you need to:
- Add support for higher‑order derivatives (second‑order, Hessians, Jacobians)
- Enable differentiation through loops, conditionals, and recursion
- Optimize gradient computation for GPU or distributed execution
- Integrate with existing neural‑network libraries (QNN, ONN, SNN)
- Implement novel AD techniques (checkpointing, source‑to‑source transformation)
When NOT to use this skill
- For simple forward‑mode AD that already works (use the existing
autodiffmodule) - For changes unrelated to differentiation (e.g., adding new math functions)
- For performance tuning of non‑AD code
Inputs required
- Specification of the AD extension (which order, which differentiation mode)
- Mathematical definitions of the new derivative operators
- Expected performance characteristics and memory usage
- Integration points with the existing AD infrastructure
Workflow
-
Understand the current AD implementation
- Read
stdlib/autodiff/andself‑hosted/check/autodiff.sio(if present) - Review the AD test suite under
tests/autodiff/ - Examine the intermediate representation used for gradient computation
- Read
-
Design the extension
- Decide whether to extend forward‑mode, reverse‑mode, or both
- Determine how higher‑order derivatives will be represented (nested tangents, Hessian matrices)
- Plan for control‑flow differentiation (e.g., taping loops, unrolling)
- Consider memory/performance trade‑offs
-
Implement the core differentiation logic
- Add new AD operators to the appropriate standard library modules
- Extend the compiler's AD lowering passes if necessary
- Ensure correctness for edge cases (discontinuities, non‑differentiable points)
-
Integrate with GPU and parallel execution
- If targeting GPU, modify the GPU backend to support AD primitives
- Add parallel gradient accumulation for distributed settings
-
Write tests
- Add unit tests for each new differentiation operator
- Include numerical correctness tests against finite‑difference approximations
- Test differentiation through control‑flow constructs
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.
- 9d ago First seen · 131 lines · 50 tokens per session scan A 1af7e4cf5d58
automatic-differentiation-advanced is a skill published in the GitHub repository Sounio-lang/sounio (6 stars, last pushed today), licensed Apache-2.0. It adds 50 tokens to every session and 1,181 once invoked, about $0.0003 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 2026-08-31.
Other skills, from other repositories
vera-language
Write programs in the Vera programming language. Use when asked to write, edit, debug, or review Vera code (.vera files). Vera is a statically typed, purely functional language with algebraic effects, mandatory contracts, and typed slot references (@T.n) instead of variable names.
vera-language
Write programs in the Vera programming language. Use when asked to write, edit, debug, or review Vera code (.vera files). Vera is a statically typed, purely functional language with algebraic effects, mandatory contracts, and typed slot references (@T.n) instead of variable names.
jupyter-notebook
Iterative Python via live Jupyter kernel (hamelnb).
migrate-oxfmt
Guide for migrating a project from Prettier or Biome to Oxfmt. Use when asked to migrate, convert, or switch a JavaScript/TypeScript project's formatter from Prettier or Biome to Oxfmt.
cuopt-numerical-optimization-api
LP, MILP, and QP (beta) with cuOpt — Python, C, and CLI. Use when the user is solving LP, MILP, or QP with any cuOpt interface.
gh-address-comments
Help address review/issue comments on the open GitHub PR for the current branch using gh CLI; verify gh auth first and prompt the user to authenticate if not logged in.