New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Icra Experiments

by brycewang-stanford in brycewang-stanford/Awesome-Journal-Skills

A skill your agent uses when designing or auditing the experimental section of an ICRA paper — real-robot versus simulation-only evidence, trial counts and success-rate reporting, task distributions…

MITAuto-check passedData & Analytics

Install Icra Experiments

skills CLI
$ npx skills add brycewang-stanford/Awesome-Journal-Skills --skill icra-experiments -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install brycewang-stanford/Awesome-Journal-Skills icra-experiments --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/brycewang-stanford/Awesome-Journal-Skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/ICRA-Skills/skills/icra-experiments .claude/skills/icra-experiments && rm -rf skills-src

Use ~/.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/

Facts

Skill name
icra-experiments
GitHub stars
1.2k
Token cost
~1.6k tokens
SKILL.md length
764 words
Files
1
Skills in repo
2,387
Repo updated
First seen
Licence
MIT

At a glance

A skill your agent uses when designing or auditing the experimental section of an ICRA paper — real-robot versus simulation-only evidence, trial counts and success-rate reporting, task distributions…

  • Works in 6 steps: Claim altitude vs evidence rung matched… → Success criterion stated verbatim; n… → Baselines: same hardware, tuned,… → …
  • Auditing the experimental section of an ICRA paper — real-robot versus simulation-only evidence
  • SKILL.md covers The evidence ladder, Trials, resets, and what a…, Baseline fairness on hardware and Sim-to-real claims, plus 5 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Icra Experiments is an agent skill from brycewang-stanford/Awesome-Journal-Skills. Use when designing or auditing the experimental section of an ICRA paper — real-robot versus simulation-only evidence, trial counts and success-rate reporting, task distributions and resets, baseline fairness on shared hardware, sim-to-real transfer claims, failure-mode analysis, and the statistics robotics reviewers actually expect.

Its SKILL.md is about 1.6k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Data & Analytics, covering Statistics. The repository describes itself as: Journal-specific Claude Code/Codex skill packs covering mainstream journals — AER, QJE, Nature, Cell, 管理世界, 经济研究 & 200+ more — your fast track to getting published. ｜ 覆盖主流期刊的… The licence is MIT.

When your agent uses it

  • Auditing the experimental section of an ICRA paper — real-robot versus simulation-only evidence
  • Trial counts and success-rate reporting
  • Task distributions and resets
  • Baseline fairness on shared hardware

Example prompts

  • “/icra-experiments”

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

  1. Claim altitude vs evidence rung matched for every abstract claim?
  2. Success criterion stated verbatim; n visible in every table/caption?
  3. Baselines: same hardware, tuned, includes strong classical option?
  4. Sim-vs-real deltas reported numerically where transfer is claimed?
  5. Failure taxonomy present, counts summing to total trials?
  6. Statistics: intervals suited to small n; one load-bearing test?

What it can do on your machine

Read from SKILL.md and the folder at commit 932eb23. 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

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Icra Experiments loads about 1.6k tokens when it runs. Until then it costs about 88 tokens; SKILL.md has 764 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~88
When it runs · the whole SKILL.md, loaded when a task matches
~1.6k

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 brycewang-stanford/Awesome-Journal-Skills at commit 932eb23, republished under its MIT licence (© brycewang-stanford). 764 words, ~1,625 tokens.

Download SKILL.mdSave it as .claude/skills/icra-experiments/SKILL.md (or your agent's skills folder).
name
icra-experiments
description
Use when designing or auditing the experimental section of an ICRA paper — real-robot versus simulation-only evidence, trial counts and success-rate reporting, task distributions and resets, baseline fairness on shared hardware, sim-to-real transfer claims, failure-mode analysis, and the statistics robotics reviewers actually expect.

ICRA Experiments

Experimental evidence is where ICRA reviews are won and lost. The community's core suspicion is the cherry-picked demo: one lucky run, filmed once, presented as capability. The experiments section exists to prove the demo was not luck.

The evidence ladder

Climb as high as the claim requires; state the rung explicitly in the paper.

RungEvidenceSupports claims like
1Simulation only, single environment"the formulation is feasible"
2Simulation, randomized dynamics/scenes"the method is robust in sim"
3Real robot, controlled lab task"works on hardware"
4Real robot, varied objects/terrains/subjects"generalizes physically"
5Extended/field deployment, uncontrolled conditions"works in the world"

Simulation-only papers are not banned at ICRA, but a rung-2 evidence base cannot carry rung-4 language. The mismatch between claim altitude and evidence rung is the single most cited weakness in robotics reviews. If hardware is unreachable, scope the claims to simulation and say why the sim is trustworthy (validated dynamics, established benchmark, physics-accurate contact).

Trials, resets, and what a success rate means

  • Pre-register the protocol privately: number of trials, success criterion, and abort rule before collecting headline numbers; report the criterion verbatim in the paper ("success = object lifted 10 cm and held 5 s").
  • Trial counts: 5 trials per condition is anecdote, 10-25 is typical ICRA practice, more when variance is high. Report n everywhere, including captions.
  • Resets are part of the method. If a human re-tees the object between trials, say so; automated-reset claims are different capability claims.
  • No silent trial deletion. "Excluding two trials due to hardware fault" is fine when disclosed with cause; undisclosed exclusion is fabrication.
  • Report dispersion, not just means: min/max or standard deviation over trials, and per-condition breakdowns rather than one pooled percentage.

Baseline fairness on hardware

Hardware baselines are expensive, so reviewers scrutinize the shortcuts:

  • Run baselines on the same platform, same task instances, same sensor stream — a baseline run in simulation against a method run on hardware is not a comparison, it is two anecdotes.
  • Tune the baseline with comparable effort and say what that effort was; "the baseline used default parameters" invites the fatal review sentence "the baseline was not tuned."
  • Include the naive-but-strong baseline (a well-engineered classical planner or PID stack). Beating only learned rivals while a 1990s controller would have matched the result is a known embarrassment pattern at robotics venues.
  • When a competitor's code is closed, reimplement and mark results "reimplementation," or compare on published numbers with the setup delta stated.

Sim-to-real claims

If the paper trains in simulation and deploys on hardware, quantify the gap, never just bridge it rhetorically:

  • Report the same metric in sim and real side by side; the delta is a result.
  • Name the randomization/adaptation mechanism and ablate it — "domain randomization" as an unexamined incantation draws fire.
  • Zero-shot vs fine-tuned transfer must be labeled; ten minutes of on-robot fine-tuning is legitimate but is a different claim.
Show full SKILL.md (285 more words)Show less

Failure analysis as a first-class section

ICRA reviewers trust papers that dissect their failures more than papers that report none. Budget for a failure table:

text
Failure taxonomy (example structure)
  F1 perception: slip not detected (4/50 trials) — cause: torque noise
     under 0.3 N·m threshold; mitigation: none in current sensing
  F2 planning: recovery pose in collision (2/50) — cause: stale octomap
  F3 hardware: gripper timeout (1/50) — excluded (disclosed), watchdog fixed
Reported: 43/50 success (86%); all 7 failures categorized above.

A reviewer reading this knows the team understands its own system. The video attachment should show at least one failure case for the same reason (see icra-supplementary).

Ablations with hardware budgets

Full ablation grids are a simulation luxury. The hardware-honest pattern:

  • Run the complete grid in simulation; run on hardware only the two or three ablations that test the paper's central mechanism.
  • Say exactly this in the paper — reviewers accept the economics when the choice of which ablations earned hardware time is justified.
  • Never present a simulation ablation in a way that lets a reader mistake it for a hardware result; label every table row's regime.

Statistics that fit robotics sample sizes

  • With n = 10-25 trials per condition, prefer exact binomial confidence intervals for success rates over normal approximations.
  • Paired designs (same object set for method and baseline) plus a paired test beat unpaired comparisons at the same n.
  • Do not run twelve significance tests on six pages; pick the one comparison the claim rests on and treat the rest descriptively.
  • Learning components: report seeds and per-seed outcomes for training, then hold the deployed policy fixed across hardware trials so trial variance is physical, not training noise.

Pre-deadline experiment audit

  1. Claim altitude vs evidence rung matched for every abstract claim?
  2. Success criterion stated verbatim; n visible in every table/caption?
  3. Baselines: same hardware, tuned, includes strong classical option?
  4. Sim-vs-real deltas reported numerically where transfer is claimed?
  5. Failure taxonomy present, counts summing to total trials?
  6. Statistics: intervals suited to small n; one load-bearing test?

Output format

text
[Evidence rung] <1-5> vs claim altitude <1-5> — mismatch: <y/n>
[Trial audit] n per condition, criterion stated, exclusions disclosed
[Baseline audit] hardware-matched? tuned? classical baseline present?
[Transfer audit] sim/real deltas numeric? adaptation labeled?
[Failure analysis] taxonomy present? counts reconcile?
[Single weakest table/figure] <which and why>

© brycewang-stanford, 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 ICRA-Skills/skills/icra-experiments of brycewang-stanford/Awesome-Journal-Skills.

Open the folder on GitHubat commit 932eb23

Compare with similar skills

Icra Experiments 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.

Icra Experiments compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Icra Experiments this skillbrycewang-stanford/Awesome-Journal-Skills1.2k—~1.6kAutomated safety check: PassMIT
Sandbox Benchvercel/next.js143k—~4.1kAutomated safety check: PassMIT
Statistical Analysisspacering-net/codeg3.8k4 repos~5kAutomated safety check: PassMIT
StatsmodelszLanqing/codex-claude-academic-skills4.6k16 repos~4.9kAutomated safety check: PassBSD-3-Clause
Statistical Powerspacering-net/codeg3.8k2 repos~3.6kAutomated safety check: NotesMIT
AI Daily DigestvigorX777/ai-daily-digest1.6k—~1.3kAutomated safety check: PassNone

Similar skills

  • Sandbox Bench

    vercel/next.js

    Official

    Benchmark React or Next.js changes on Vercel Sandbox VMs with paired A/B statistics: react PR/commit vs base, or Next.js PR/commit vs base, measured end-to-end through the bench/render-pipeline app…

    143k GitHub stars~4.1k tokensUpdated yesterday
    Data & AnalyticsAuto-check passed

  • Statistical Analysis

    spacering-net/codeg

    Guided statistical analysis for research data - test selection, assumption checking, effect sizes, power analysis, Bayesian alternatives, and APA-formatted reporting.

    3.8k GitHub starsUsed in 4 repos~5k tokens
    Data & AnalyticsAuto-check passed

  • Statsmodels

    zLanqing/codex-claude-academic-skills

    Statistical models library for Python. An agent skill from zLanqing/codex-claude-academic-skills.

    4.6k GitHub starsUsed in 16 repos~4.9k tokens
    Data & AnalyticsAuto-check passed

  • Statistical Power

    spacering-net/codeg

    Sample-size and statistical power calculations for planning studies.

    3.8k GitHub starsUsed in 2 repos~3.6k tokens
    Data & AnalyticsAuto-check: notes

  • AI Daily Digest

    vigorX777/ai-daily-digest

    Fetches RSS feeds from 90 top Hacker News blogs (curated by Karpathy), uses AI to score and filter articles, and generates a daily digest in Markdown with Chinese-translated titles, category…

    1.6k GitHub stars~1.3k tokensUpdated 7 mo ago
    Data & AnalyticsAuto-check passed

  • Agent Session Monitor

    higress-group/higress

    Real-time agent conversation monitoring - monitors Higress access logs, aggregates conversations by session, tracks token usage.

    9.5k GitHub stars~3.3k tokensUpdated 4 days ago
    Data & AnalyticsAuto-check passed

More from brycewang-stanford/Awesome-Journal-Skills

All 2,387 skills in this repo

  • Aaag Data Analysis

    brycewang-stanford/Awesome-Journal-Skills

    A skill your agent uses when running and reporting the analysis for an Annals of the American Association of Geographers manuscript — spatial statistics and modeling, remote-sensing accuracy, or…

    1.2k GitHub stars~1.3k tokensUpdated 11 days ago
    Auto-check passed

  • Aaag Literature Positioning

    brycewang-stanford/Awesome-Journal-Skills

    A skill your agent uses when positioning an Annals of the American Association of Geographers manuscript in the literature — engaging geographic scholarship across the relevant area and the…

    1.2k GitHub stars~1.3k tokensUpdated 11 days ago
    Auto-check passed

  • Aaag Rebuttal

    brycewang-stanford/Awesome-Journal-Skills

    A skill your agent uses when responding to an Annals of the American Association of Geographers decision letter (major/minor revision) — building a point-by-point response to the subject editor and…

    1.2k GitHub stars~1.4k tokensUpdated 11 days ago
    Auto-check passed

  • Aaag Research Design

    brycewang-stanford/Awesome-Journal-Skills

    A skill your agent uses when defending the research design of an Annals of the American Association of Geographers manuscript — spatial/quantitative analysis and GIScience, remote-sensing and…

    1.2k GitHub stars~1.4k tokensUpdated 11 days ago
    Auto-check passed

  • Aaag Review Process

    brycewang-stanford/Awesome-Journal-Skills

    A skill your agent uses when you need to understand how the Annals of the American Association of Geographers evaluates a manuscript — double-anonymous review routed through a subject editor by…

    1.2k GitHub stars~1.3k tokensUpdated 11 days ago
    Auto-check passed

  • Aaag Submission

    brycewang-stanford/Awesome-Journal-Skills

    A skill your agent uses when running the final pre-submission preflight for the Annals of the American Association of Geographers via ScholarOne Manuscripts — area/article-type selection…

    1.2k GitHub stars~1.6k tokensUpdated 11 days ago
    Auto-check passed

Categories

Questions about Icra Experiments

What does Icra Experiments do?

A skill your agent uses when designing or auditing the experimental section of an ICRA paper — real-robot versus simulation-only evidence, trial counts and success-rate reporting, task distributions…. Icra Experiments is an agent skill from brycewang-stanford/Awesome-Journal-Skills. Use when designing or auditing the experimental section of an ICRA paper — real-robot versus simulation-only evidence, trial counts and success-rate reporting, task distributions and resets, baseline fairness on shared hardware, sim-to-real transfer claims, failure-mode analysis, and the statistics robotics reviewers actually expect.

When should I use Icra Experiments?

Icra Experiments fits situations like: auditing the experimental section of an ICRA paper — real-robot versus simulation-only evidence; trial counts and success-rate reporting; task distributions and resets; baseline fairness on shared hardware.

How do I install Icra Experiments in Claude Code?

Run `npx skills add brycewang-stanford/Awesome-Journal-Skills --skill icra-experiments -a claude-code`. Or copy the skill folder (ICRA-Skills/skills/icra-experiments in brycewang-stanford/Awesome-Journal-Skills) into .claude/skills/icra-experiments in your project. Claude Code loads it when a task matches its description.

How do I install Icra Experiments in Codex?

Run `npx skills add brycewang-stanford/Awesome-Journal-Skills --skill icra-experiments -a codex`. Or copy the skill folder (ICRA-Skills/skills/icra-experiments in brycewang-stanford/Awesome-Journal-Skills) into .agents/skills/icra-experiments in your project. Codex loads it when a task matches its description.

Can I use Icra Experiments 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 brycewang-stanford/Awesome-Journal-Skills --skill icra-experiments -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/icra-experiments, .gemini/skills/icra-experiments, .github/skills/icra-experiments and .opencode/skills/icra-experiments in your project.

What does Icra Experiments need to run?

SKILL.md names no scripts, command-line tools or credentials: Icra Experiments is instructions for the agent only.

Does Icra Experiments access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Icra Experiments 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 Icra Experiments use?

Icra Experiments 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 Icra Experiments use?

About 1.6k tokens (SKILL.md is roughly 6.5k 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 Icra Experiments?

Skills that share tags, products or a category with Icra Experiments: Sandbox Bench (vercel/next.js, 143k stars), Statistical Analysis (spacering-net/codeg, 3.8k stars), Statsmodels (zLanqing/codex-claude-academic-skills, 4.6k stars) and Statistical Power (spacering-net/codeg, 3.8k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Icra Experiments?

brycewang-stanford (a GitHub user) maintains it in brycewang-stanford/Awesome-Journal-Skills, which has 1,219 GitHub stars. The repository holds 2,387 skills in this directory. The repository was last updated on September 27, 2026.

Source: brycewang-stanford/Awesome-Journal-Skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.