extend-new-decoder SKILL.md skillA
extend-new-decoder from qutone/lightstim is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add qutone/lightstim/extend-new-decoder@git:20260808.cfe5e96Pins git:20260808.cfe5e96 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- qutone/lightstim
- Path
skills/extend-new-decoder/SKILL.md- Versions recorded
- 3
- Latest version
- git:20260808.cfe5e96, committed Aug 8, 2026
- First version recorded
- Jun 3, 2026
- Size
- 10,599 bytes
- Repository stars
- 50
- License
- a license file GitHub does not identify
Integrate a new decoder into LightStim's decoder backend so it works with `SimulationPipeline` and `DecoderConfig`. Use this skill whenever the user asks to add a custom decoder, wrap a third-party decoder library (e.g. a neural decoder, a research-paper decoder), register a CPU or GPU decoder backend, build a decoder from scratch using a DEM, or extend the existing BPOSD/PyMatching/MWPF stack with a new variant.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Add a New Decoder
LightStim's decoder backend is a registry of decoder classes. Register a class
under a name + backend (`cpu` / `gpu` / `fpga`), and
`DecoderConfig(name="…", backend="…", params={…})` finds it at runtime —
everywhere: pipeline, server, notebooks, benchmarks.
Adding a decoder is always three things:
1. **One new file** in `lightstim/simulation/decoder_backend/decoders/`.
2. **One soft-import line** in `decoders/__init__.py`.
3. **One smoke test** in `tests/test_simulation_backend_quality.py`.
A runnable version of everything below (with live benchmarks) is
`tutorials/how-to-add-new-decoders.ipynb`.
## Which pattern?
| Situation | Pattern |
|---|---|
| Upstream library ships a `sinter.Decoder` | **A** — register the class directly |
| Anything else | **C** — subclass `ExternalDecoder` (recommended) |
| You need to own the unpack → decode → pack loop yourself | **B** — build from the DEM matrices |
---
## The contract (duck-typed)
The pipeline never checks `isinstance(..., sinter.Decoder)`. A decoder is any
object with:
```python
class MyDecoder: # sinter.Decoder base optional
…Read the whole file at its exact version.
How to install
mdr add qutone/lightstim/extend-new-decoder@sha256:077d662c5be4a1fcThe line at the top of this page pins the label git:20260808.cfe5e96, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_5b3qlhvhkhfzh6lc)
3 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 3 versions of this file, between Jun 3, 2026 and Aug 8, 2026, about one every 33 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260808.cfe5e96 latest | 2026-08-08 | cfe5e96 | 10,599 B | A | view · diff |
| git:20260606.819d2e0 | 2026-06-06 | 819d2e0 | 19,838 B | A | view · diff |
| git:20260603.dbc60ac | 2026-06-03 | dbc60ac | 14,404 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (10599 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
qutone/lightstim · 50 stars · license a license file GitHub does not identify · pushed 2026-10-29 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_5b3qlhvhkhfzh6lc GET https://markdownregistry.com/api/v1/resolve?ref=qutone/lightstim/extend-new-decoder GET https://markdownregistry.com/api/v1/blob/077d662c5be4a1fccd77f0098c5c89cd7c04b22a734a1e383dfa87f9f1ee080f
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.