Home / grainulation / grainulator · skills/witness/SKILL.md · GitHub

witness skillA

witness is agent-read markdown (skill) from grainulation/grainulator: Corroborate a claim against an external source URL..

Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.

What the file says

# /witness -- External corroboration

The user wants to verify a claim against a specific external source.

## Arguments

$ARGUMENTS

Expected format: `/witness <claim_id> <url> [--smart] [--mode concise|full|meta-only]`

The `--smart` flag uses Grainulator’s smart-fetch MCP tool instead of raw WebFetch. Smart mode extracts only title, description, and main content — typically 80-99% smaller, faster to read, and cached locally for 7 days.

## Persona: Fact-Checker

You are a methodical evidence auditor with neutral stance. Verify source credibility (publication date, author expertise, track record), cross-reference against conflicting data, identify outdated or single-sourced evidence. Upgrade claims if stronger evidence emerges; downgrade if contradictions appear.

## What corroboration requires

- An **independent** source — one that cites primary data and isn't circling back to the same
  original source.
- Judgement of the specific page, not the domain: a reputable site still publishes opinion,
  stale data, and sponsored content.
- **Inverse** search terms and a recorded pass count, so "no contradictions found after N
  passes" is a result rather than a shrug.
…

Read the whole file at its exact version.

How to install

Latest version
mdr add grainulation/grainulator/witness@git:20260922.e702f49
Exact content
mdr add grainulation/grainulator/witness@sha256:9257d493b2106bdd

Pin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_swx4nuc6zx3jdg6u.svg)](https://markdownregistry.com/a/art_swx4nuc6zx3jdg6u)

1 badge views in 30 days

Versions

versioncommittedcommitsizeaudit
git:20260922.e702f49 latest2026-09-22 e702f49 5,634 BA view · diff
git:20260908.608fac02026-09-08 608fac0 4,800 BA view · diff
git:20260419.c471fc02026-04-19 c471fc0 3,244 BA view · diff
git:20260418.d0330b72026-04-18 d0330b7 3,234 BA view · diff
git:20260407.e152df62026-04-07 e152df6 2,676 BA view

Audit of the latest version

A  17 of 17 checks passed. Deterministic, no model, 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 (5634 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

grainulation/grainulator · 86 stars · license MIT · pushed 2026-09-22 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_swx4nuc6zx3jdg6u
GET https://markdownregistry.com/api/v1/resolve?ref=grainulation/grainulator/witness
GET https://markdownregistry.com/api/v1/blob/9257d493b2106bddcca8b76f91df3c90d1b8708c0db1ca3aa263d45df5038ea9

Your agent 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.

More from grainulation/grainulator

AGENTS.md agents
grainulation/grainulator · AGENTS.md
git:20260908.8af756a · audit A · 86 stars
llms.txt@packages/analytics/site llmstxt
grainulation/grainulator · packages/analytics/site/llms.txt
git:20260908.608fac0 · audit A · 86 stars
AGENTS.md@packages/evidence agents
grainulation/grainulator · packages/evidence/AGENTS.md
git:20260908.608fac0 · audit A · 86 stars
llms.txt@packages/evidence/site llmstxt
grainulation/grainulator · packages/evidence/site/llms.txt
git:20260908.608fac0 · audit A · 86 stars
llms.txt@packages/exports/site llmstxt
grainulation/grainulator · packages/exports/site/llms.txt
git:20260908.608fac0 · audit A · 86 stars
llms.txt@packages/legacy-cli/site llmstxt
grainulation/grainulator · packages/legacy-cli/site/llms.txt
git:20260908.608fac0 · audit A · 86 stars
llms.txt@packages/memory/site llmstxt
grainulation/grainulator · packages/memory/site/llms.txt
git:20260908.608fac0 · audit A · 86 stars
llms.txt@packages/orchestration/site llmstxt
grainulation/grainulator · packages/orchestration/site/llms.txt
git:20260908.608fac0 · audit A · 86 stars
bean skill
grainulation/grainulator · packages/runtime/skills/bean/SKILL.md · Run a task as a recursive convergence loop, not a fixed checklist: investigate the most decisive open question, record…
git:20260908.608fac0 · audit A · 86 stars
llms.txt@packages/shared/site llmstxt
grainulation/grainulator · packages/shared/site/llms.txt
git:20260908.608fac0 · audit A · 86 stars
analytics skill
grainulation/grainulator · skills/analytics/SKILL.md · Sprint analytics — type distributions, stale claims, velocity, prediction scoring
git:20260922.e702f49 · audit A · 86 stars
blind-spot skill
grainulation/grainulator · skills/blind-spot/SKILL.md · Structural gap analysis -- find what the sprint has not considered.
git:20260922.e702f49 · audit A · 86 stars

Every file in grainulation/grainulator

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