model-bakeoff SKILL.md skillA
This file is byte-identical to the first copy the registry indexed. Same content hash, so the same checks give both the same grade.
model-bakeoff from narendawar/narens-claude-toolkit is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add narendawar/narens-claude-toolkit/model-bakeoff@git:20261010.895cd4cPins git:20261010.895cd4c in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- narendawar/narens-claude-toolkit
- Path
plugins/model-bakeoff/skills/model-bakeoff/SKILL.md- Versions recorded
- 3
- Latest version
- git:20261010.895cd4c, committed Oct 10, 2026
- First version recorded
- Oct 10, 2026
- Size
- 6,272 bytes
- Repository stars
- 4
- License
- MIT
Use when the user wants to pick the cheapest model that still works for a custom subagent, asks whether Haiku or Sonnet would do, or wants to re-check an agent's model after new models ship (for example 'which model should this agent use?', 'could this run on Haiku?', 'bake-off my reviewer agent').
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
# Model Bake-off ## Overview A custom subagent runs on whatever its `model` field says, usually chosen by feel. This skill measures instead: it runs the agent's test cases on Haiku, Sonnet and Opus through `claude -p`, applies a stated pass rule, recommends the cheapest model that passes, and only after the user says yes writes it into the agent's `model` field. It runs only when the user asks. It never invents a number: the script does all the arithmetic, and it states what the runs do and do not show. ## The script `scripts/bakeoff.py` in this skill's directory (use the base directory shown when the skill loads). Python 3 and the standard library only; use `python3`, or `python` / `py -3` if that is what works. - `agents [--json]` lists custom agents with their model, tools and whether they are read-only. - `check-cases --cases FILE` validates a cases file. - `estimate --agent NAME --cases FILE [--models LIST] [--runs N]` shows the number of runs, the ceiling cost, and whether an opt-in is needed. It calls nothing. …
Read the whole file at its exact version.
How to install
mdr add narendawar/narens-claude-toolkit/model-bakeoff@sha256:12d5bb10e6506c3eThe line at the top of this page pins the label git:20261010.895cd4c, 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_muja2u647k5zmo4d)
1 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 3 versions of this file, all committed on Oct 10, 2026.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261010.895cd4c latest | 2026-10-10 | 895cd4c | 6,272 B | A | view · diff |
| git:20261010.2ae1ce6 | 2026-10-10 | 2ae1ce6 | 5,772 B | A | view · diff |
| git:20261010.e9f4360 | 2026-10-10 | e9f4360 | 5,407 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 (6272 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
narendawar/narens-claude-toolkit · 4 stars · license MIT · pushed 2026-10-20 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_muja2u647k5zmo4d GET https://markdownregistry.com/api/v1/resolve?ref=narendawar/narens-claude-toolkit/model-bakeoff GET https://markdownregistry.com/api/v1/blob/12d5bb10e6506c3e3d49400b82d345647f83d167024806d265b5184ac44c54d6
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.
More from narendawar/narens-claude-toolkit
Every file in narendawar/narens-claude-toolkit