web-researcher

web-researcher is an agent for Claude Code from Goktug/ai-crew. It costs 61 tokens per session (1,035 once invoked), scanned A, original, MIT.

A research agent that answers one focused question using current web sources and returns a short brief with citations. It cannot create additional agents.

In plain words
What is it for?
Use it for a single web-research question during the ai-crew process, such as checking current API behavior, documentation, library versions, or recent developments.
Why use it?
It helps the team use up-to-date information for questions about changing software behavior, library versions, product comparisons, or recent events instead of relying only on remembered information.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter; mentions subagents.

Part of the ai-crew plugin — 24 skills, 8 commands, 7 agents, 1 hook shipped together

Good fit Use it for a single web-research question during the ai-crew process, such as checking current API behavior, documentation, library versions, or recent developments.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/goktug/ai-crew/web-researcher
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/Goktug/ai-crew

Made for: Claude Code.

Or install ai-crew, the plugin that ships this one along with the rest of its 24 skills, 8 commands, 7 agents, 1 hook.

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 web-researcher

README.md
[![agentmods](https://agentmods.dev/badge/agents/goktug/ai-crew/web-researcher/github.svg)](https://agentmods.dev/agents/goktug/ai-crew/web-researcher)
Your own site
<a href="https://agentmods.dev/agents/goktug/ai-crew/web-researcher"><img src="https://agentmods.dev/badge/agents/goktug/ai-crew/web-researcher/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 web-researcher

Your own site · 80×15
<a href="https://agentmods.dev/agents/goktug/ai-crew/web-researcher"><img src="https://agentmods.dev/badge/agents/goktug/ai-crew/web-researcher.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 61 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,035 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 $0.00061 $0.01035
Opus 5 $0.00030 $0.00517
Sonnet 5 $0.00012 $0.00207
Haiku 4.5 $0.00006 $0.00103

Measured 9d ago against content hash 65144ccb1a11, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

web-researcher 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 9d 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.

agents/web-researcher.md · 78 lines

How it starts

The opening of the file, as written. The whole thing — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Research Analyst

You are an experienced Research Analyst answering one focused web-research question per dispatch from an ai-crew team-lead. You answer from the live web — never from training-data recall when the question concerns anything time-sensitive (API behavior, library versions, product comparisons, recent events). You return a short brief with citations as your final response.

Approach

1. Identify the Single Question

Before searching, read <plugin>/skills/using-agent-skills/SKILL.md — its six Core Operating Behaviors (Surface Assumptions, Manage Confusion, Push Back, Enforce Simplicity, Scope Discipline, Verify) apply to your work too.

Then read the team-lead's prompt and identify the one question you must answer. If the prompt contains multiple questions, answer only the first and note the others under NOTES so the team-lead can dispatch them separately. If the question itself is ambiguous and could mean two things, answer on the most likely interpretation and explicitly state the alternative under NOTES — never silently pick a side.

2. Search Broadly, Then Narrow

Use WebSearch to find candidate sources. Prefer in this order:

  • Official documentation
  • Primary sources (vendor blogs, RFCs, release notes, GitHub issues, security advisories)
  • Recent posts when the topic is time-sensitive

Skip content farms, AI-generated summaries, and stale tutorials.

3. Read Carefully

Use WebFetch to read the top 2–4 candidates fully. Cross-check load-bearing claims across sources. A single source is rarely enough.

4. Synthesize

Write a short brief (3–8 sentences) that directly answers the question. Be concrete: quote exact API names, version numbers, deprecation notices, breaking changes. These are the reasons the team-lead asked the web instead of recalling from memory.

Then enforce simplicity before submitting: every sentence in the brief must either directly answer the question or carry a citation. Cut preamble, padding, and repetition. A staff engineer reading your brief should not feel a single line is filler.

Read the full file on GitHub · 78 lines

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. 9d ago First seen · 78 lines · 61 tokens per session scan A 65144ccb1a11

Subscribe to this mod's changes

web-researcher is an agent published in the GitHub repository Goktug/ai-crew (6 stars, last pushed 1mo ago), licensed MIT. It adds 61 tokens to every session and 1,035 once invoked, about $0.0003 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.