skill-selection

skill-selection is a cursor rule for Cursor from mamahoos/dot-files. It costs 552 tokens per session, scanned A, original, MIT.

A set of rules for choosing only the skills needed for a task and loading them in a defined order.

In plain words
What is it for?
It helps select skills for small edits, software changes, unclear requirements, and task-specific areas such as user interfaces.
Why use it?
It reduces unnecessary context and prevents agents from applying unrelated instructions.

Cursor rule for Cursor

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 rules/mamahoos/dot-files/skill-selection
Clone the repo
git clone --depth 1 https://github.com/mamahoos/dot-files

Made for: Cursor.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for skill-selection

README.md
[![agentmods](https://agentmods.dev/badge/rules/mamahoos/dot-files/skill-selection.svg)](https://agentmods.dev/rules/mamahoos/dot-files/skill-selection)
Your own site
<a href="https://agentmods.dev/rules/mamahoos/dot-files/skill-selection"><img src="https://agentmods.dev/badge/rules/mamahoos/dot-files/skill-selection.svg" alt="Measured on agentmods" height="20"></a>
Per session 552 This file is loaded in full into every session.
When invoked 552 The same file — it is already loaded in full.
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 $0.00552 $0.00552
Opus 5 $0.00276 $0.00276
Sonnet 5 $0.00110 $0.00110
Haiku 4.5 $0.00055 $0.00055

Measured yesterday against content hash 840af50b1c1e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

skill-selection 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 yesterday.

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.

home/.cursor/rules/skill-selection.mdc · 55 lines

How it starts

The opening of the file, as written. The whole thing — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Skill selection

using-agent-skills is the catalog. This rule only says how many skills to load. Do not copy that tree here.

Budget

Load the smallest set that covers the task. Do not walk the full discovery tree. Do not read a SKILL.md unless that skill is in the set for this turn.

Tiny work

Obvious one-file, low-risk edits: git-workflow-and-versioning only. Skip the rest; one line why if a lifecycle step is skipped.

Non-trivial software change

These seven, in order (skip a step only with a one-line why):

  1. spec-driven-development
  2. planning-and-task-breakdown
  3. incremental-implementation
  4. test-driven-development
  5. debugging-and-error-recovery — only when something is already broken
  6. code-review-and-quality
  7. git-workflow-and-versioning

Unclear intent before a spec: interview-me or idea-refine, then step 1.

Extras — signal only

After the spine, add extras only for a matching branch of the using-agent-skills discovery tree (example: UI work → frontend-ui-engineering, not every Build skill).

If the job is not in that tree, match the task to a description: in home/.cursor/skills/*/SKILL.md. Do not open every file; search descriptions or use the name the user typed. User /name or an explicit skill name wins for this turn.

Same job, two names

Prefer the core-lifecycle skill unless the user invoked the other:

Job Use Not unless named
TDD test-driven-development tdd
Review code-review-and-quality code-review
Debug debugging-and-error-recovery diagnosing-bugs

Slots the discovery tree does not have

  • Hard-to-reverse stack, boundary, storage, or deploy: architecture-decision-making (then documentation-and-adrs only to record).
  • GitHub issues, PRs, labels, merge: github-workflow (in addition to git-workflow-and-versioning).
  • User looking for an installable skill: find-skills.

If an extra skill is used, one short note: which and why.

Read the full file on GitHub · 55 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. yesterday First seen · 55 lines · 552 tokens per session scan A 840af50b1c1e

Subscribe to this mod's changes

skill-selection is a cursor rule published in the GitHub repository mamahoos/dot-files (4 stars, last pushed yesterday), licensed MIT. It adds 552 tokens to every session, about $0.0028 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 2026-09-04.