shellbench-research-runbook

A runbook for conducting full ShellBench benchmark campaigns. ShellBench is a test suite for comparing coding agents on shell-based tasks across several agent tools and models.

In plain words
What is it for?
It covers remote benchmark execution, fixed software and model versions, qualification runs, repeated trials, trace and usage checks, judge routing, artifact export, and result retention.
Why use it?
It provides repeatable rules for running, checking, recording, and auditing benchmark results. This helps distinguish a one-off local score from a documented research campaign.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/openclaw/shellbench/shellbench-research-runbook
Any agent
npx skills add openclaw/shellbench --skill shellbench-research-runbook
Clone the repo
git clone --depth 1 https://github.com/openclaw/shellbench

Made for: Claude Code, Codex.

Per session 76 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,124 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00076 $0.01124
Opus 5 $0.00038 $0.00562
Sonnet 5 $0.00015 $0.00225
Haiku 4.5 $0.00008 $0.00112

Measured 3d ago against content hash 7c327fcaaae0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

shellbench-research-runbook 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 3d 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.

.agents/skills/shellbench-research-runbook/SKILL.md · 111 lines

How it starts

The opening of the file, as written. The whole thing — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ShellBench Research Runbook

Use this skill for a real benchmark campaign, not a one-off local score.

Read references/runbook.md before provisioning machines. It is the normative campaign contract and contains the commands, gates, artifact schema, and recovery rules.

Non-negotiable gates

  1. Use remote Crabbox AWS beasts for benchmark execution. Never run scored trials on the operator laptop.
  2. Pin one public-task commit, runner commit or patch hash, provider model ID, harness version, reasoning level, and judge route for the whole campaign.
  3. Run one r0 qualification for every distinct harness and model-family route, using exactly ten pinned representative tasks. Do not start full-suite jobs until model identity, real traces, tools, usage, judge routing, and artifact export pass.
  4. Retain and audit every r0, but force it out of leaderboard scoring. Qualify with independent full-suite repetitions r1 through r3. After a clean audit, add r4 through r6; the research result is six total repetitions.
  5. Run every provider-supported non-maximum reasoning level. Never label a reasoning level as tested unless the route applies it and the trace or proxy evidence proves it. Record unsupported levels instead of fabricating them.
  6. Use gpt-5.6-sol at high as the default judge. Keep the judge alias, credentials, logs, and identity audit separate from the agent route.
  7. Start checkpointing after the first completed trial and continue at least every ten minutes or ten new results. Verify each local archive before it counts.
  8. Upload every verified checkpoint and final archive to the private S3 prefix from SHELLBENCH_TRACE_S3_URI. Never put bucket names or credentials in git, PR text, public logs, or generated reports.
  9. A run is not research-clean when traces are missing, observed model identity differs from the request, reasoning is unproven, coverage is incomplete, or infrastructure failures dominate.

Read the full file on GitHub · 111 lines

Files

What ships with it

2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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.

  1. 3d ago First seen · 111 lines · 76 tokens per session scan A 7c327fcaaae0

Subscribe to this mod's changes

shellbench-research-runbook is a skill published in the GitHub repository openclaw/shellbench (138 stars, last pushed 5d ago), licensed MIT. It adds 76 tokens to every session and 1,124 once invoked, about $0.0004 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 2026-08-30.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens

chat-perf

Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.

microsoft/vscode · 51 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens