# Skill Token Budget Linter

> Lint skill descriptions, body length, and reference bulk against token budgets. Use when tightening standing-context cost. NOT for DCP tuning or RTK hooks.

- Skill: `wyattowalsh/skill-token-budget-linter` (Agent Skill, multi-file: 5 files)
- Install (CLI): `npx skillmds@latest add wyattowalsh/skill-token-budget-linter`
- Raw SKILL.md: https://api.skillmd.com/api/skills/wyattowalsh/skill-token-budget-linter/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Finance & Business
- License: MIT
- Author: wyattowalsh (https://skillmd.com/u/wyattowalsh)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/wyattowalsh/skill-token-budget-linter

---


# Skill Token Budget Linter

Measure standing-context pressure for skill assets before they ship.

**Scope:** Static token-budget linting only. Does not change OpenCode DCP, RTK, or harness hook policy.

## Dispatch

| $ARGUMENTS | Action |
|------------|--------|
| `<name>` | Lint one skill directory |
| `repo` / `--all` | Lint every `skills/*/SKILL.md` |
| `--strict` | Fail on warning-tier budgets, not just hard limits |
| Empty | Show budget table and example commands |

## Budget Table

| Surface | Soft warn | Hard fail |
|---------|-----------|-----------|
| `description` | 200 chars | 1024 chars |
| body (below frontmatter) | 400 lines | 500 lines |
| single reference file | 300 lines | 500 lines |

## Workflow

```bash
uv run python skills/skill-token-budget-linter/scripts/lint_budget.py <name>
uv run python skills/skill-token-budget-linter/scripts/lint_budget.py --all
```

## Validation Contract

```bash
uv run python skills/skill-token-budget-linter/scripts/check.py
```