Skip to content
Agentmart
Submit a tool

classifying-tasks

Use when receiving any new work request, before writing code or choosing an approach — especially multi-item requests, vague one-liners, "quick" asks, or work…

Skill by Roflochinsky

IndexedMIT
Write a review

Chalk Labs

Need help deploying classifying-tasks?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/roflochinsky-classifying-tasks

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Use when receiving any new work request, before writing code or choosing an approach — especially multi-item requests, vague one-liners, "quick" asks, or work under deadline pressure.

Use cases: writing

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/classifying-tasks

What's inside

Skill name
classifying-tasks
Description
Use when receiving any new work request, before writing code or choosing an approach — especially multi-item requests, vague one-liners, "quick" asks, or work under deadline pressure.
License
MIT
Instructions
40 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Classifying Tasks

## Overview

The cost of being wrong — not the size of the ask — determines how much process a task needs. Classify BEFORE acting. A "small column" that changes a contract is heavy; a 500-line copy update is trivial.

**A deadline never downgrades the tier.** Time pressure changes what scope you negotiate, never which gates you skip.

## How to classify

1. If the project's CLAUDE.md defines its own tiers/dispatcher — follow it. This skill is the fallback and the default for projects without one.
2. Classify **each item separately** when a request contains several asks. One message ≠ one tier.
3. Check signals top-down, **first match wins**. Unsure between tiers → take the higher one or ask ONE question.

| Tier | Signals (any one) | Route |
|---|---|---|
| **Heavy** | New feature with its own business logic; ≥2 layers with a new/changed contract (e.g. DB + API + UI); touches a gated decision: auth/session, schema strategy, money, metric/status semantics, public API, RBAC/visibility; destructive or hard-to-reverse operation | **delivering-heavy-features** skill |
| **Standard** | Real behavior change, one layer or a thin slice, no gated decision | Brief clarification if ambiguous → tracker issue → short plan if non-obvious (superpowers:writing-plans) → TDD → verify (superpowers:verification-before-completion) |
| **Trivial** | Copy/label/locale text, single-spot style or cosmetic tweak, comment/docs typo | Do directly, surgical diff; verify only what's relevant (typecheck/lint) |

A bug fix is classified by its **fix**, not its symptom: a fix that changes a formula or contract is Heavy. Any debugging starts with superpowers:systematic-debugging regardless of tier.

## State it out loud

Before starting, tell the user the classification in one line:
"Item 2 is heavy (DB+API+UI, new contract) — running the heavy pipeline; items 1 and 3 are trivial/standard."

This gives the user a chance to veto before process diverges.

## Red flags — you are rationalizing

| Thought | Reality |
|---|---|
| "Deadline is tight, I'll run a compressed pipeline" | Compression = silently making the user's product decisions. Negotiate scope down instead: ship fewer items at full discipline. |
| "It's just one small column" | Columns come with contracts, defaults, migrations, and consumers. Classify by blast radius, not line count. |
| "The user is unavailable, so I'll skip the questions" | Unavailability defers gates, it doesn't remove them. See delivering-heavy-features for how to proceed safely. |
| "I'll classify after I look at the code" | Looking is fine; coding before classifying is not. |
| "The whole batch is basically one task" | Batches hide a heavy item behind two trivial ones. Per-item, always. |

From Roflochinsky/workshopskills, MIT.

More from Roflochinsky/workshopskills

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Jun 16, 2026 (4 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from Roflochinsky

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get classifying-tasks running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying classifying-tasks?

CallGet help
Cite this page

classifying-tasks skill. Agent Mart. https://agentmart.dev/skills/roflochinsky-classifying-tasks (last updated 2026-10-24)