task-creator

A workflow for creating submission-ready SkillsBench tasks, which are benchmark tasks used to evaluate coding-agent skills. It guides the author from the initial idea through task files, tests, reference answers, and validation.

In plain words
What is it for?
Use it to propose and scaffold a task, prepare its environment and tests, write an answer key, select skills, and run validation.
Why use it?
It provides the required structure and checks so a task is complete, testable, and suitable for benchmark evaluation.

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/benchflow-ai/benchflow/task-creator
Any agent
npx skills add benchflow-ai/benchflow --skill task-creator
Clone the repo
git clone --depth 1 https://github.com/benchflow-ai/benchflow

Made for: Claude Code, Codex.

Per session 104 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,360 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 2 findings. Scan, not verified.
Origin 97% copy Near-identical to another mod 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 $0.00104 $0.04360
Opus 5 $0.00052 $0.02180
Sonnet 5 $0.00021 $0.00872
Haiku 4.5 $0.00010 $0.00436

Measured 2d ago against content hash 3785158f65d7, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

task-creator scanned grade B with 2 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 2d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (scripts/preflight.sh), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

CODEX_MODEL=gpt-5.5 # adjust per ~/.codex/config.toml

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

Bundle frozen inputs (CSVs, xlsx) in `environment/`. For tasks that need internet (research-track), declare the required network and environment policy in `task.md` frontmatter and prefer Playwright over `urllib` for any
Origin

This is a copy

97% identical to task-creator — 11 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

.agents/skills/task-creator/SKILL.md · 239 lines

How it starts

The opening of the file, as written. The whole thing — 239 lines — stays where its author put it; the contents beside it link to each section on GitHub.

SkillsBench Task Authoring

Repo context. This skill lives in the benchflow repo but authors tasks for benchflow-ai/skillsbench. Unqualified references below — CONTRIBUTING.md, MAINTAINER.md, docs/*.md, tasks/<task-id>/ — mean files in skillsbench, not in this repo. Work from a skillsbench checkout.

Build a task that scores well on the task principles. Two artifacts when you're done: a directory under tasks/<task-id>/ that bench tasks check accepts, and a PR description that maps cleanly to the PR template.

Workflow

1. propose      → one-paragraph proposal, gut-check against the proposal rubric
2. scaffold     → bench tasks init, plus the native task.md layout below
3. task.md      → frontmatter + human-written, outcome-focused prompt body
4. environment  → Dockerfile + bundled inputs; do NOT bake skills
5. tests        → 4–10 test functions, parametrize for bulk; check formulas AND values
6. oracle       → human-written reference solution that derives answers
7. skills       → 2–3 generalizable skills (or reuse existing ones from /tasks/*/environment/skills/)
8. validate     → bench tasks check + oracle eval (must reach 1.0)
9. self-review  → invoke task-review skill on the local path
10. agent runs  → Opus 4.8 / latest Codex with and without skills
11. submit      → PR with the table the template asks for

Each step is described below. Skip a step only if the rubric says it's optional for your track (research / multimodal). Skipping verification will get the PR rejected.

Step 1 — Propose

Before writing files, write a four-bullet proposal answering the proposal-stage rubric:

  • What's the task? One paragraph.
  • Who does this in real life? Job, domain, why someone pays for it.
  • Why do skills help? What domain knowledge is non-obvious without a skill?
  • How would you verify it? Specific output files, deterministic tests.

Read the full file on GitHub · 239 lines

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. 2d ago First seen · 239 lines · 104 tokens per session scan B 3785158f65d7

Subscribe to this mod's changes

task-creator is a skill published in the GitHub repository benchflow-ai/benchflow (335 stars, last pushed 2d ago), licensed Apache-2.0. It adds 104 tokens to every session and 4,360 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it B with 2 findings (reads agent configuration directories, makes network calls). It is 97% identical to task-creator, differing in 11 lines, and is treated as a copy.

Related

Other skills, from other repositories

lastlight-evals

Scaffold, configure and run a Last Light EVALS workspace — the harness that runs Last Light's real workflows against a mocked GitHub and grades them deterministically. Use when the user wants to "set up / scaffold Last Light Evals", "create an evals workspace or instance", "run evals", "compare models", or author new…

nearform/lastlight · 185 tokens

hotpath_init

Configure hotpath profiling in a Rust project. Adds the hotpath dependency with feature-gated setup, instruments main with hotpath::main, functions with measure/measureall, and wraps channels, mutexes, rwlocks, streams, futures, reqwest clients, axum routers and byte-level I/O with hotpath macros. Use when the user…

pawurb/hotpath-rs · 88 tokens

review-pr

Walk Sean through an incoming waku-agent PR or issue and present it his way — four fixed sections: what this is, why it matters, how HE can test it with you as copilot, and are we ready to merge / reply / close and why. Use whenever Sean asks to look at, test, triage, or decide on a pull request or an issue, and…

ShenSeanChen/waku-agent · 93 tokens

compose-graphics

Advanced Compose visuals - Material 3 Expressive motion physics, AGSL shaders (Android 13+), Canvas/DrawScope generative, graphicsLayer effects.

Jwuthri/Tracely-ai · 36 tokens

canvas-generative

Algorithmic and generative art with Canvas 2D - particles, flow fields, noise, fractals, L-systems.

Jwuthri/Tracely-ai · 30 tokens

writing-bench-task-judge

Use when writing or modifying checkgoals() / getanswer() / App check methods in benchenv/task/, or when reviewing a draft task's judge correctness. Triggers include adding a new task, editing a judge method, or diagnosing a judge false-positive/negative.

Purewhiter/mobilegym · 68 tokens