Skip to content

Skills / budget

budget

Strict low-cost mode for the rest of the session, with tighter reading limits and terse replies. Slash command only. Run it when you want the agent to spend as few tokens as possible.

OfficialAuditA
Latest
v1.0.0
Aisle
workflow
Works with
Claude Code, Codex, Cursor
Stars
2

Install

Install budget

npx @agentsaisle/cli add budget

Installs into your project's skills folder. Add --global to install it for every project.

Audit

A
Critical
0
High
0
Medium
0
Low
0

3 files · audited 2026-10-29 · rules 1.0.0

No findings.

What this skill can use

Tools

  • Read
  • Grep
  • Glob

Scripts

None

Domains mentioned

None

Automated checks found no known risky patterns. This is not a guarantee of safety. How audits work · Read the rules

Show this grade in a README

The badge updates when the registry is rebuilt. Paste this Markdown into the README:

[![aisle audit](https://img.shields.io/endpoint?url=https%3A%2F%2Fagentsaisle-vercel.vercel.app%2Fbadges%2Fbudget.json)](https://agentsaisle-vercel.vercel.app/skills/budget)

SKILL.md

Rendered from commit 4036c95

Budget

Use when

  • The user runs /budget to turn on strict mode for the rest of the session.

Don't use when

  • The task needs broad exploration. Tell the user and ask whether to turn strict mode off.

Steps

  1. Confirm in one line that strict mode is on.
  2. Before reading anything, search. Find the exact place with grep or a symbol search, then read only that range.
  3. Keep every read narrow: at most 4 files per step and at most 120 lines per read.
  4. If the scope is unclear, ask one short question instead of exploring.
  5. After two failed attempts at the same problem, stop and report what you tried.
  6. Stay in strict mode until the user says to stop.

Budget

  • At most 4 files and 120 lines per read. Never re-read a file you just wrote or edited.
  • Do not restate the request, list what you are about to do, or explain code you did not change.
  • The final reply is at most 10 lines: what changed, what you ran, what is left.

Output

  • Terse answers. Code and results first, no preamble, no recap of work the user can see.

Done when

  • Each reply is within the limits above.
  • The user ended strict mode, or the session ended.

Version history

VersionCommitInstall
1.0.0latest4036c95add budget

Report this skillAgent Aisle on GitHub