verify-before-claiming SKILL.md skillA

verify-before-claiming from emeraldleaf/okl is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add emeraldleaf/okl/verify-before-claiming@git:20260831.7854149

Pins git:20260831.7854149 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
emeraldleaf/okl
Path
src/okl/scaffold/claude/skills/verify-before-claiming/SKILL.md
Versions recorded
1
Latest version
git:20260831.7854149, committed Aug 31, 2026
Size
3,227 bytes
Repository stars
4
License
MIT

Use before stating that work is done, a test passes, a build is green, a bug is fixed, a file exists, or any factual claim about the state of the code or the world. Requires running the check and reading its real output before asserting the result — evidence precedes assertion, always.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# Verify before claiming

A claim about state — "the tests pass," "the file is created," "the endpoint returns
404," "19 tests green" — is only worth as much as the check behind it. Stating a result
you *expect* as if it were a result you *observed* is the single most common way
AI-assisted work goes fast, fluent, plausible, and wrong.

This skill has one rule: **run the check, read the output, then make the claim — in that
order.** If you have not run the check this turn, you do not have the result; you have a
guess. Say "I expect" for a guess and "I confirmed" only for an observation.

## When this fires

- About to write "done", "fixed", "passing", "green", "works", "created", "verified".
- About to put a count, a status, or a value into prose, a commit message, a README, a
  PR description, or a report.
- About to tell a human that a thing is true about the code or the system.

## The procedure

1. **Name the check.** What single command or observation would make this claim true or
   false? (`pytest -q`, `ls path`, `curl -s -o /dev/null -w '%{http_code}'`, `git status`.)
2. **Run it this turn.** Not "I ran it earlier" — state drifts; a file moved, a kernel
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add emeraldleaf/okl/verify-before-claiming@sha256:29aea1e1562523dc

The line at the top of this page pins the label git:20260831.7854149, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_lolxqwm724ek5agq.svg)](https://markdownregistry.com/a/art_lolxqwm724ek5agq)

2 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 1 version of this file, committed on Aug 31, 2026.

versioncommittedcommitsizeaudit
git:20260831.7854149 latest2026-08-31 7854149 3,227 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (3227 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

emeraldleaf/okl · 4 stars · license MIT · pushed 2026-11-07 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_lolxqwm724ek5agq
GET https://markdownregistry.com/api/v1/resolve?ref=emeraldleaf/okl/verify-before-claiming
GET https://markdownregistry.com/api/v1/blob/29aea1e1562523dcf39b3f4bbdfb2687b4664524dea187472d9720b780020ec1

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from emeraldleaf/okl

encoding-loop SKILL.md skill
emeraldleaf/okl · .claude/skills/encoding-loop/SKILL.md · Use when you discover a rule, antipattern, or lesson worth keeping ("we should never write this again" / "we should…
git:20261026.04d23e4 · audit A · 4 stars
AGENTS.md AGENTS.md file
emeraldleaf/okl · AGENTS.md
git:20261107.ff72c19 · audit A · 4 stars
CLAUDE.md CLAUDE.md file
emeraldleaf/okl · CLAUDE.md
git:20261107.ff72c19 · audit A · 4 stars
encoding-loop SKILL.md skill
emeraldleaf/okl · src/okl/scaffold/claude/skills/encoding-loop/SKILL.md · Use when you discover a rule, antipattern, or lesson worth keeping ("we should never write this again" / "we should…
git:20261026.04d23e4 · audit A · 4 stars
CLAUDE.md@src/okl/scaffold/root CLAUDE.md file
emeraldleaf/okl · src/okl/scaffold/root/CLAUDE.md
git:20260930.f1f54de · audit A · 4 stars

Every file in emeraldleaf/okl

Other files named verify-before-claiming

verify-before-claiming SKILL.md skill
leonardleroy/i-dont-believe-you · .claude/skills/verify-before-claiming/SKILL.md · Use for any task that will end in a claim about tests, the build, or whether the work is done: making tests pass…
git:20260924.425ec48 · audit A · 18 stars
verify-before-claiming SKILL.md skill
leonardleroy/i-dont-believe-you · .cursor/skills/verify-before-claiming/SKILL.md · Use for any task that will end in a claim about tests, the build, or whether the work is done: making tests pass…
git:20260924.425ec48 · audit A · 18 stars
verify-before-claiming SKILL.md skill
leonardleroy/i-dont-believe-you · .codex/skills/verify-before-claiming/SKILL.md · Use for any task that will end in a claim about tests, the build, or whether the work is done: making tests pass…
git:20260924.425ec48 · audit A · 18 stars
verify-before-claiming SKILL.md skill
leonardleroy/i-dont-believe-you · .agents/skills/verify-before-claiming/SKILL.md · Use for any task that will end in a claim about tests, the build, or whether the work is done: making tests pass…
git:20260924.425ec48 · audit A · 18 stars
verify-before-claiming SKILL.md skill
leonardleroy/i-dont-believe-you · .opencode/skills/verify-before-claiming/SKILL.md · Use for any task that will end in a claim about tests, the build, or whether the work is done: making tests pass…
git:20260924.425ec48 · audit A · 18 stars
verify-before-claiming SKILL.md skill
leonardleroy/i-dont-believe-you · skills/verify-before-claiming/SKILL.md · Use for any task that will end in a claim about tests, the build, or whether the work is done: making tests pass…
git:20260924.425ec48 · audit A · 18 stars
verify-before-claiming SKILL.md skill
yoavshmariahu/thinker · .agents/skills/verify-before-claiming/SKILL.md · Check a claim about thinker's own cost or behaviour against ground truth before asserting it: the API's reported usage…
git:20261104.ecb0077 · audit A · 16 stars

Browse by kind, by grade A, or by owner.