web-scout

web-scout is an agent for Claude Code from ikotelkin/claude-skills. It costs 57 tokens per session (404 once invoked), scanned A, original, MIT.

A low-cost web research worker that scans pages, listings, or other online sources and records findings in files. It returns a short digest, while the full results include source URLs and the dates they were checked.

In plain words
What is it for?
Use it to compare listings, check many pages, collect sourced facts, and save research findings for another agent to review.
Why use it?
It handles repetitive searching and fetching so the main agent does not have to spend its time on volume-heavy research.

Agent for Claude Code

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

Part of the frugal-subagents plugin — 1 skill, 2 agents, 1 hook shipped together

Good fit Use it to compare listings, check many pages, collect sourced facts, and save research findings for another agent to review.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/ikotelkin/claude-skills/web-scout
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/ikotelkin/claude-skills

Made for: Claude Code.

Or install frugal-subagents, the plugin that ships this one along with the rest of its 1 skill, 2 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-scout

README.md
[![agentmods](https://agentmods.dev/badge/agents/ikotelkin/claude-skills/web-scout/github.svg)](https://agentmods.dev/agents/ikotelkin/claude-skills/web-scout)
Your own site
<a href="https://agentmods.dev/agents/ikotelkin/claude-skills/web-scout"><img src="https://agentmods.dev/badge/agents/ikotelkin/claude-skills/web-scout/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-scout

Your own site · 80×15
<a href="https://agentmods.dev/agents/ikotelkin/claude-skills/web-scout"><img src="https://agentmods.dev/badge/agents/ikotelkin/claude-skills/web-scout.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 57 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 404 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 $0.00057 $0.00404
Opus 5 $0.00028 $0.00202
Sonnet 5 $0.00011 $0.00081
Haiku 4.5 $0.00006 $0.00040

Measured 4d ago against content hash d177d129ce17, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

web-scout 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 4d 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.

skills/frugal-subagents/agents/web-scout.md · 22 lines

What it actually says

You are a research scout. Your caller is a more expensive model that has delegated the volume work to you; your job is to gather, not to decide.

Operate like this:

  • Work from the brief only. If it is ambiguous, pick the most literal reading and say so in the digest — don't ask, you have no one to ask.
  • Write findings to the file(s) named in the brief as you go (append after every few sources), so nothing is lost if you are cut off by a rate limit or turn cap. If the brief names no file, write scout-findings.md in the working directory and say so.
  • Prefer breadth first: one pass over all targets with the cheapest check, then depth only where the first pass found something.
  • Record provenance: every fact carries its URL and the date you saw it. Distinguish what a page states from what you infer.
  • Stop when the brief's question is answered, the target list is exhausted, or you have made ~30 fetches — whichever comes first. Say which.

Your final message is a digest of at most ~15 lines: what you covered, the 3–7 findings that matter, what you could not reach, and the path(s) of the file(s) holding the full material. No transcripts, no page dumps — the caller reads the files if it needs detail.

You cannot spawn agents. If the task turns out to need several parallel workers, finish your own slice and say in the digest how you would split the rest.

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. 4d ago First seen · 22 lines · 57 tokens per session scan A d177d129ce17

Subscribe to this mod's changes

web-scout is an agent published in the GitHub repository ikotelkin/claude-skills (3 stars, last pushed 5d ago), licensed MIT. It adds 57 tokens to every session and 404 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 2026-09-08.

Related

Other agents, from other repositories

plan-implementer

Implement code based on a specific plan or task description. This agent is designed for focused, incremental implementation work where there is a clear specification or plan to execute. Use when there are well-defined implementation tasks from a feature plan.

mhattingpete/claude-skills-marketplace · 50 tokens

docs-manager

Use this agent when you need to manage technical documentation, establish implementation standards, analyze and update existing documentation based on code changes, write or update Product Development Requirements (PDRs), organize documentation for developer productivity, or produce documentation summary reports. This…

vanducng/skills · 84 tokens

web-researcher

Searches the web for a specific research query and returns a compressed finding with cohort-diverse citations and source-type annotations. Use when the parent recipe needs current-web information — especially fast-moving landscape/tooling topics where model priors can't be trusted.

ethanaubuchon/dossier-tradecraft · 54 tokens

researcher

Use this agent when you need to conduct comprehensive research on software development topics, including investigating new technologies, finding documentation, exploring best practices, or gathering information about plugins, packages, and open source projects. This agent excels at synthesizing information from…

vanducng/skills · 356 tokens

inventory-seeder

Read scattered findings from legacy audit folders and generate canonical 00-inventory.md rows. Used by /aidex-audit migrate after folders have been moved.

yacb2/aidex · 35 tokens

conventions-auditor

Runs the aidex-conventions validator (validate.py) against the project's .context/ and reports violations as aidex findings.

yacb2/aidex · 30 tokens