XanSAFE

Intelligent Skill routing and workflow orchestration for AI agents — +21.12 pp reward, −29.6% tokens on SkillsBench with DeepSeekV4Flash-VE.

The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.

---
name: xan
description: High-performance CSV processing with xan CLI for large tabular datasets, streaming transformations, and low-memory pipelines.
---

# xan

Use this skill when tabular work involves large CSV-like files and shell-oriented data pipelines.

## When to Use

- CSV/TSV files are large (size or rows) and Python/pandas style processing is too heavy.
- You need fast CLI operations such as `filter`, `sort`, `dedup`, `groupby`, `frequency`, `join`.
- You want low-memory streaming workflows in terminal pipelines.

## Boundaries

- Not for workbook semantics (`.xlsx` formula/layout preservation) — use `xlsx` or `excel-analysis`.
- Not for ML model training pipelines — use `data-ml` skills.
- This skill is a thin CLI wrapper; it does not implement custom parsing logic itself.

## Script

- Path: `scripts/xan.ps1`

### Examples

```powershell
powershell -ExecutionPolicy Bypass -File "$env:USERPROFILE\.codex\skills\xan\scripts\xan.ps1" count data.csv
powershell -ExecutionPolicy Bypass -File "$env:USERPROFILE\.codex\skills\xan\scripts\xan.ps1" filter "score > 0.8" data.csv
powershell -ExecutionPolicy Bypass -File "$env:USERPROFILE\.codex\skills\xan\scripts\xan.ps1" groupby category "sum(amount) as total" sales.csv
```

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

What the source does

Filesystem

none-observed

Network

none-observed

Shell

none-observed

Dependencies

pinned

Secrets in source

none-found

Findings (0)

No findings outside the package's declared scope.

Gates applied: no_behavioural_pass.