Defuddle
kepano/obsidian-skills
Uses the Defuddle CLI to pull clean, readable Markdown, JSON or metadata from web pages, stripping navigation, ads and clutter to save tokens.
How to reproduce and grade a bench task or A/B arm for moss capability work
MITAuto-check passed
Install Bench Repro
$ npx skills add D-Robotics/moss --skill bench-repro -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install D-Robotics/moss bench-repro --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/D-Robotics/moss.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.moss/skills/bench-repro .claude/skills/bench-repro && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "bench-repro" agent skill from https://github.com/D-Robotics/moss/tree/main/.moss/skills/bench-repro into .claude/skills/bench-repro/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "bench-repro", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/D-Robotics/moss/tree/main/.moss/skills/bench-reproType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add D-Robotics/moss --skill bench-repro -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install D-Robotics/moss bench-repro --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/D-Robotics/moss.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.moss/skills/bench-repro .agents/skills/bench-repro && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "bench-repro" agent skill from https://github.com/D-Robotics/moss/tree/main/.moss/skills/bench-repro into .agents/skills/bench-repro/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "bench-repro", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add D-Robotics/moss --skill bench-repro -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install D-Robotics/moss bench-repro --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/D-Robotics/moss.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.moss/skills/bench-repro .cursor/skills/bench-repro && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "bench-repro" agent skill from https://github.com/D-Robotics/moss/tree/main/.moss/skills/bench-repro into .cursor/skills/bench-repro/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "bench-repro", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/D-Robotics/moss.git --path .moss/skills/bench-repro--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add D-Robotics/moss --skill bench-repro -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install D-Robotics/moss bench-repro --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/D-Robotics/moss.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.moss/skills/bench-repro .gemini/skills/bench-repro && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "bench-repro" agent skill from https://github.com/D-Robotics/moss/tree/main/.moss/skills/bench-repro into .gemini/skills/bench-repro/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "bench-repro", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install D-Robotics/moss bench-reproInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add D-Robotics/moss --skill bench-repro -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/D-Robotics/moss.git skills-src && mkdir -p .github/skills && cp -r skills-src/.moss/skills/bench-repro .github/skills/bench-repro && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "bench-repro" agent skill from https://github.com/D-Robotics/moss/tree/main/.moss/skills/bench-repro into .github/skills/bench-repro/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "bench-repro", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add D-Robotics/moss --skill bench-repro -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install D-Robotics/moss bench-repro --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/D-Robotics/moss.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.moss/skills/bench-repro .opencode/skills/bench-repro && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "bench-repro" agent skill from https://github.com/D-Robotics/moss/tree/main/.moss/skills/bench-repro into .opencode/skills/bench-repro/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "bench-repro", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
bench-repro- GitHub stars
- 143
- Token cost
- ~288 tokens
- SKILL.md length
- 113 words
- Files
- 1
- Skills in repo
- 3
- Repo updated
- First seen
- Licence
- MIT
At a glance
How to reproduce and grade a bench task or A/B arm for moss capability work
- Calls npm; needs MOSS_BENCH_API_KEY
What it does
Bench Repro is an agent skill from D-Robotics/moss. How to reproduce and grade a bench task or A/B arm for moss capability work
Its SKILL.md is about 290 tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It works with npm. The licence is MIT.
Example prompts
- “/bench-repro”
Requirements
- A credential in MOSS_BENCH_API_KEY
What it can do on your machine
Read from SKILL.md and the folder at commit 58e7f61. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
npm
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use npm, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names these keys or tokens, usually read from environment variables:
MOSS_BENCH_API_KEY
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Bench Repro loads about 288 tokens when it runs. Until then it costs about 22 tokens; SKILL.md has 113 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from D-Robotics/moss at commit 58e7f61, republished under its MIT licence (© D-Robotics). 113 words, ~288 tokens.
Download SKILL.mdSave it as
.claude/skills/bench-repro/SKILL.md (or your agent's skills folder).- name
- bench-repro
- description
- How to reproduce and grade a bench task or A/B arm for moss capability work
- when
- running benchmarks, investigating a bench regression, or collecting A/B evidence
Reproducing bench evidence
- Internal bench:
npm run bench -- --task <id> --samples <n> --label <label>(provider viaMOSS_BENCH_API_KEY,--model,--base-url; temp=0; approval=never). Results land inbench/results/<label>/(never committed). - A/B:
npm run bench:ab -- <engine> --samples <n>forbest-of-n/reasoning-high/goal-loop/model-routing; the verdict json lands inbench/results/ab-<engine>-<stamp>.json. - Noise discipline: compare only same-SHA runs; use
npm run bench:noisefor the noise band; a delta inside the band is not evidence. - SWE-bench (external board):
npm run bench:swe -- --samples 2 --label <label>produces predictions; grade withnpm run bench:swe -- --eval --label <label>. Instance subset is locked inbench/boards/swebench-instances.json— never edit it mid-study. - A baseline must be collected from a pinned build (
--dist bench/.cache/<snapshot>/dist), not from a dirty tree.
© D-Robotics, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in .moss/skills/bench-repro of D-Robotics/moss.
Open the folder on GitHubat commit 58e7f61
Compare with similar skills
Bench Repro next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Bench Repro this skillD-Robotics/moss | 143 | — | ~288 | Automated safety check: Pass | MIT | |
| Defuddlekepano/obsidian-skills | 49k | 12 repos | ~208 | Automated safety check: Pass | MIT | |
| Vercel Deploybytedance/deer-flow | 83k | 10 repos | ~797 | Automated safety check: Pass | MIT | |
| MCP Server BuildershareAI-lab/learn-claude-code | 78k | 5 repos | ~1.2k | Automated safety check: Pass | MIT | |
| Knap Markdown Templateskepano/obsidian-skills | 49k | 2 repos | ~986 | Automated safety check: Pass | MIT | |
| Install Anti-Slop Oxlint Rulesdmmulroy/anti-slop | 5.3k | 1 repos | ~2.2k | Automated safety check: Pass | MIT |
Similar skills
Vercel Deploy
bytedance/deer-flow
Deploys a project to Vercel with one script and no login, then returns a live preview URL and a claim link for moving the deployment into your own Vercel account.
MCP Server Builder
shareAI-lab/learn-claude-code
Walks through building MCP servers in Python or TypeScript that expose tools, resources and prompts to Claude, with templates, registration and testing.
Knap Markdown Templates
kepano/obsidian-skills
Renders Markdown notes from Knap templates and JSON data on the command line, including notes built from Defuddle web page output.
Install Anti-Slop Oxlint Rules
dmmulroy/anti-slop
Installs, updates or migrates the vendored anti-slop Oxlint plugin in a repository, keeping local rule changes and the plugin's license and provenance files.
Nx Run Tasks
nomcopter/react-mosaic
Helps with running tasks in an Nx workspace. An agent skill from nomcopter/react-mosaic.
More from D-Robotics/moss
Patch Discipline
D-Robotics/moss
How to make code changes in the moss repo — read-before-edit, minimal diffs, verify with real runs
Worktree Parallel
D-Robotics/moss
How to run writable sub-agents in isolated git worktrees and merge their lease patches
Related
Works with
Questions about Bench Repro
What does Bench Repro do?
How to reproduce and grade a bench task or A/B arm for moss capability work. Bench Repro is an agent skill from D-Robotics/moss.
How do I install Bench Repro in Claude Code?
Run `npx skills add D-Robotics/moss --skill bench-repro -a claude-code`. Or copy the skill folder (.moss/skills/bench-repro in D-Robotics/moss) into .claude/skills/bench-repro in your project. Claude Code loads it when a task matches its description.
How do I install Bench Repro in Codex?
Run `npx skills add D-Robotics/moss --skill bench-repro -a codex`. Or copy the skill folder (.moss/skills/bench-repro in D-Robotics/moss) into .agents/skills/bench-repro in your project. Codex loads it when a task matches its description.
Can I use Bench Repro in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add D-Robotics/moss --skill bench-repro -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/bench-repro, .gemini/skills/bench-repro, .github/skills/bench-repro and .opencode/skills/bench-repro in your project.
What does Bench Repro need to run?
Going by SKILL.md and its folder, Bench Repro needs the command-line tools its instructions call (npm) and credentials named MOSS_BENCH_API_KEY. Our summary lists: A credential in MOSS_BENCH_API_KEY.
Does Bench Repro access the network?
SKILL.md contains no URLs. Its commands use npm, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Is Bench Repro safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Bench Repro use?
Bench Repro is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Bench Repro use?
About 288 tokens (SKILL.md is roughly 1.2k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Bench Repro?
Skills that share tags, products or a category with Bench Repro: Defuddle (kepano/obsidian-skills, 49k stars), Vercel Deploy (bytedance/deer-flow, 83k stars), MCP Server Builder (shareAI-lab/learn-claude-code, 78k stars) and Knap Markdown Templates (kepano/obsidian-skills, 49k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Bench Repro?
D-Robotics (a GitHub organization) maintains it in D-Robotics/moss, which has 143 GitHub stars. The repository holds 3 skills in this directory. The repository was last updated on October 2, 2026.
Source: D-Robotics/moss on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.