agentmods.dev

corpus-kb: Command for Claude Code

.claude/commands/corpus-nli-classify.md

corpus-nli-classify is a command for Claude Code from moliver28/corpus-kb. It costs 7 tokens per session (307 once invoked), scanned A, original, MIT.

A zero-shot text-classification command that asks a language model whether text supports supplied labels. Zero-shot means it can propose labels without being trained on examples specific to your dataset.

In plain words
What is it for?
Use it for pilot labeling of text and review the returned support scores before relying on the labels.
Why use it?
It provides initial category proposals without building a separate classifier, but results can change with wording, label descriptions, negation, and overlapping categories.

Command for Claude Code

Written for Claude Code: argument-hint in frontmatter.

This is moliver28/corpus-kb's own configuration. It tells Claude Code how to work on corpus-kb itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything corpus-kb configures →

not rated 1repo today A scan 7 tokens original MIT
View source ↗ moliver28/corpus-kb
Reuse

Borrowing it

Nothing to install: this file belongs to moliver28/corpus-kb. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/moliver28/corpus-kb/master/.claude/commands/corpus-nli-classify.md
Clone the repo
git clone --depth 1 https://github.com/moliver28/corpus-kb

Made for: Claude Code.

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 corpus-nli-classify

README.md
[![agentmods](https://agentmods.dev/badge/commands/moliver28/corpus-kb/corpus-nli-classify/github.svg)](https://agentmods.dev/commands/moliver28/corpus-kb/corpus-nli-classify)
Your own site
<a href="https://agentmods.dev/commands/moliver28/corpus-kb/corpus-nli-classify"><img src="https://agentmods.dev/badge/commands/moliver28/corpus-kb/corpus-nli-classify/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for corpus-nli-classify

Your own site · 80×15
<a href="https://agentmods.dev/commands/moliver28/corpus-kb/corpus-nli-classify"><img src="https://agentmods.dev/badge/commands/moliver28/corpus-kb/corpus-nli-classify.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 7 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 307 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00007 $0.00307
Opus 5.5 $0.00003 $0.00123
Sonnet 5.5 $0.00001 $0.00061
Haiku 4.5 $0.00001 $0.00031

Measured 13d ago against content hash 64612019faf3, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

corpus-nli-classify 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 13d ago.

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.

.claude/commands/corpus-nli-classify.md · 29 lines

What it actually says

corpus-nli-classify — analysis method card

The axis walk (before you run)

  1. State the research question in one sentence.
  2. Match the output family this op serves: labels (method 18).
  3. Check the inputs it needs: raw_text, llm (an Ollama NLI model).
  4. Budget the calibration effort (moderate) and autonomy (automated_with_audit).
  5. Read the cautions. Run. Then do the validation step.

Run

corpus-kb nli-classify --label <label> ... [--chunk-id <id> ...] [--model <tag>]
  • Cautions: wording, lexical overlap, and syntax can swing the support judgment. Label descriptions shape the results.
  • Validation: test paraphrased labels, negation, and overlapping labels. Tune thresholds on labeled data.
  • Claim only: pilot label proposals with support scores.
  • A guard path (disabled, empty, unavailable) returns a reason and a remedy, never a guessed result. Exit codes: 0 answered, 1 failed, 3 unavailable.
  • All flags: corpus-kb nli-classify --help
  • Method rows: docs/ANALYSIS-METHODS.md (the register is the source).
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. 13d ago First seen · 29 lines · 7 tokens per session scan A 64612019faf3

Subscribe to this mod's changes

corpus-nli-classify is a command published in the GitHub repository moliver28/corpus-kb (1 stars, last pushed today), licensed MIT. It adds 7 tokens to every session and 307 once invoked, about $0.0000 per session on Opus 5.5. 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 2026-10-24.

Analytics and metricsData and AI
Related

corpus-nli-classify compared with commands from other repositories

corpus-nli-classify next to 6 commands from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
corpus-nli-classifymoliver28/corpus-kbCommand1todayMITA7Claude Code
bootmicrosoft/PromptKitCommand1139d agoMITA14GitHub Copilot
music-suno-promptfrankxai/Starlight-Intelligence-SystemCommand10yesterdayMITA23Claude Code
audit-promptOwl-Listener/ai-design-skillsCommand1995mo agoMITA13Claude Code
laravel-ai-sdkjpcaparas/superpowers-laravelCommand1574mo agoMITA27Claude Code
vlm-ocr-evaluationscdenney/open-science-skillsCommand69yesterdaynoneA0Claude Code
dare-llm-integrationdewtech-technologies/dare-methodCommand53mo agoMITA0Claude Code