verify

verify is a command for Claude Code from Syedomershah99/being-human. It costs 16 tokens per session (408 once invoked), scanned A, original, MIT.

A writing check that compares text with the user’s own previous writing to see whether it plausibly sounds like them.

In plain words
What is it for?
It helps test pasted text or a file, measure how typical it is against the user’s writing, and identify wording patterns that make it seem less like them.
Why use it?
It addresses a different problem from checking for common AI writing patterns: whether the text matches the person’s usual authorship style.

Command for Claude Code

Written for Claude Code: allowed-tools in frontmatter.

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is python3 scripts/verify.py "$ARGUMENTS" --in .being-human/.

Part of the being-human plugin — 1 skill, 5 commands, 1 hook, 1 MCP server shipped together

Good fit It helps test pasted text or a file, measure how typical it is against the user’s writing, and identify wording patterns that make it seem less like them.

Compare 6 commands from other repositories ↓
Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo
git clone --depth 1 https://github.com/Syedomershah99/being-human
agentmods
npx agentmods add commands/syedomershah99/being-human/verify

Made for: Claude Code.

Or install being-human, the plugin that ships this one along with the rest of its 1 skill, 5 commands, 1 hook, 1 MCP server.

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 verify

README.md
[![agentmods](https://agentmods.dev/badge/commands/syedomershah99/being-human/verify.svg)](https://agentmods.dev/commands/syedomershah99/being-human/verify)
Your own site
<a href="https://agentmods.dev/commands/syedomershah99/being-human/verify"><img src="https://agentmods.dev/badge/commands/syedomershah99/being-human/verify.svg" alt="Measured on agentmods" height="20"></a>
Per session 16 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 408 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 $0.00016 $0.00408
Opus 5 $0.00008 $0.00204
Sonnet 5 $0.00003 $0.00082
Haiku 4.5 $0.00002 $0.00041

Measured 7d ago against content hash 4f77211f0156, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

verify 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 7d 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.

commands/verify.md · 45 lines

What it actually says

Run the impostor test on: $ARGUMENTS

python3 scripts/verify.py "$ARGUMENTS" --in .being-human/

If they pasted text rather than a path, write it to a scratch file and test that.

This asks a different question from /being-human:check. That one looks for AI tells. This one asks whether the text is a plausible sample of this person, using authorship distance against a null resampled from their own writing at the same length.

Reading the result:

  • under 75th percentile — indistinguishable from their writing
  • 75–90 — within range
  • 90–97 — unusual for them
  • over 97 — reads as a different hand

The percentile is what matters, not the raw delta, which has no absolute scale. A 60th percentile means the draft is more typical of them than 40% of what they actually wrote — that's a pass, not a near-miss.

The what pushed it out list is the actionable part. It names the exact function words that are over- or under-used against their baseline, with the rates. Fixing two or three of those usually moves the percentile more than any amount of rewriting by feel.

Two things to be straight about:

Both checks are needed. A draft can clear this one and still be full of tells, because authorship distance keys on function-word distribution and generic prose often has unremarkable grammar. Run /being-human:check too.

And below about 40 words the number is noise — there isn't enough text for word frequencies to mean anything. Say so rather than reporting a percentile as if it were solid.

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. 7d ago First seen · 45 lines · 16 tokens per session scan A 4f77211f0156

Subscribe to this mod's changes

verify is a command published in the GitHub repository Syedomershah99/being-human (0 stars, last pushed 24d ago), licensed MIT. It adds 16 tokens to every session and 408 once invoked, about $0.0001 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 2026-08-31.