researcher-runner SKILL.md skillA

researcher-runner from zhiming-dev/roster is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add zhiming-dev/roster/researcher-runner@git:20260713.7c58db4

Pins git:20260713.7c58db4 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
zhiming-dev/roster
Path
researcher-agent/researcher-runner/SKILL.md
Versions recorded
2
Latest version
git:20260713.7c58db4, committed Jul 13, 2026
First version recorded
Jun 7, 2026
Size
2,574 bytes
Repository stars
12
License
MIT

Procedure the Researcher agent follows to answer a question with web search + page fetch: plan queries, search, open the promising pages, cross-check, and return a source-cited synthesis. Use when a task needs current or external facts.

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

# Researcher runner

The runbook for the **Researcher** sub-agent. The runtime injects the concrete
`SEARCH:` / `FETCH:` tool protocols into your system prompt; this skill is the *method*
you apply.

## Procedure

1. **Decompose the question.** Identify the specific facts you need (e.g. index level +
   % change for three indexes, plus the top drivers). One search rarely covers all of
   it.
2. **Search to locate, FETCH to read.** A search result is a lead: when one looks like it
   holds the answer, FETCH its url and take the fact from the page itself — never from
   the snippet. If you already know where the data lives (an official stats page, FRED,
   a filing), skip the search and FETCH the source directly. Budget: 3 searches and 6
   fetches per turn — spend fetches before extra searches.
3. **Cross-check.** When a number matters, try to see it in more than one fetched source.
   Note disagreements rather than silently picking one.
4. **Synthesize with citations.** Map each material claim to the URL it came from. Keep
   it tight.
5. **Declare gaps.** Anything you could not verify goes in an explicit "could not verify"
   line. Never paper over a gap with a plausible guess.
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add zhiming-dev/roster/researcher-runner@sha256:a13f841fe0568b9c

The line at the top of this page pins the label git:20260713.7c58db4, 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_n5uy5xxpo3lfzp5l.svg)](https://markdownregistry.com/a/art_n5uy5xxpo3lfzp5l)

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

Versions

The registry has recorded 2 versions of this file, between Jun 7, 2026 and Jul 13, 2026, about one every 36 days.

versioncommittedcommitsizeaudit
git:20260713.7c58db4 latest2026-07-13 7c58db4 2,574 BA view · diff
git:20260607.31b1c632026-06-07 31b1c63 1,795 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 (2574 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

zhiming-dev/roster · 12 stars · license MIT · pushed 2026-10-12 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_n5uy5xxpo3lfzp5l
GET https://markdownregistry.com/api/v1/resolve?ref=zhiming-dev/roster/researcher-runner
GET https://markdownregistry.com/api/v1/blob/a13f841fe0568b9cbb565dc7fdb46067998bd1abc576d05c0a2548617602fae0

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 zhiming-dev/roster

AGENTS.md AGENTS.md file
zhiming-dev/roster · AGENTS.md
git:20260630.0601754 · audit A · 12 stars
CLAUDE.md CLAUDE.md file
zhiming-dev/roster · CLAUDE.md
git:20260627.f5b0ce5 · audit A · 12 stars
coder-runner SKILL.md skill
zhiming-dev/roster · coder-agent/coder-runner/SKILL.md · Implement one task on a feature branch. Use when: the Planner has dispatched a `task_assignment` to the Coder role…
git:20260607.31b1c63 · audit A · 12 stars
e2e-runner SKILL.md skill
zhiming-dev/roster · e2e-agent/e2e-runner/SKILL.md · Run E2E test suites using playwright-cli. Use when: running test suites, executing smoke tests, running E2E/bug-bash…
git:20260607.31b1c63 · audit B · 12 stars
planner-runner SKILL.md skill
zhiming-dev/roster · planner-agent/planner-runner/SKILL.md · The Planner's operating procedure. Use when: turning a principal goal into a Plan; revising a plan after rejection…
git:20260607.cffeb43 · audit A · 12 stars
qa-runner SKILL.md skill
zhiming-dev/roster · qa-agent/qa-runner/SKILL.md · Validate and fact-check another agent's output against evidence and the task's success criteria. Use when: verifying a…
git:20260713.7c58db4 · audit A · 12 stars
reviewer-runner SKILL.md skill
zhiming-dev/roster · reviewer-agent/reviewer-runner/SKILL.md · Review a diff against a task's success criteria. Use when: a `task_assignment` for role `reviewer` lands; you need to…
git:20260607.31b1c63 · audit A · 12 stars
approval-gate SKILL.md skill
zhiming-dev/roster · shared/approval-gate/SKILL.md · Classify a proposed action's risk tier (T0–T4), apply the gate policy, and (if required) summarize the proposal into a…
git:20260607.cffeb43 · audit A · 12 stars
council SKILL.md skill
zhiming-dev/roster · shared/council/SKILL.md · Run a cross-model deliberation: pose the same question to multiple model bindings (e.g. Claude, GPT, Gemini), have them…
git:20260607.31b1c63 · audit A · 12 stars
provenance SKILL.md skill
zhiming-dev/roster · shared/provenance/SKILL.md · Append a structured event to the run's provenance log. Every agent calls this for every meaningful state transition…
git:20260607.cffeb43 · audit A · 12 stars
example-skill SKILL.md skill
zhiming-dev/roster · starter-packs/portable-core-template/.github/skills/example-skill/SKILL.md · Example skill scaffold. Replace this with a real, project-specific procedural playbook. The description must clearly…
git:20260626.368d116 · audit A · 12 stars
AGENTS.md@starter-packs/portable-core-template AGENTS.md file
zhiming-dev/roster · starter-packs/portable-core-template/AGENTS.md
git:20260626.368d116 · audit A · 12 stars

Every file in zhiming-dev/roster

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