web-researcher

web-researcher is an agent for Claude Code from Wirasm/prp. It costs 35 tokens per session (514 once invoked), scanned A, original, MIT.

A research assistant that checks current facts, software APIs, platform behavior, and technical claims against external sources. It prefers official documentation, specifications, source code, release notes, and maintainers' discussions.

In plain words
What is it for?
Use it when a technical decision depends on current external information or a precise product capability. It researches the relevant version and returns evidence, applicability, conflicts, and remaining gaps.
Why use it?
It reduces the risk of relying on outdated or unsupported information. It distinguishes documented facts, reproduced behavior, inferences, conflicts, and unanswered questions, with citations for important claims.

Agent for Claude Code

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.

agentmods
npx agentmods add agents/wirasm/prp/web-researcher
Clone the repo
git clone --depth 1 https://github.com/Wirasm/prp

Made for: Claude Code.

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/wirasm/prp/web-researcher.svg)](https://agentmods.dev/agents/wirasm/prp/web-researcher)
Your own site
<a href="https://agentmods.dev/agents/wirasm/prp/web-researcher"><img src="https://agentmods.dev/badge/agents/wirasm/prp/web-researcher.svg" alt="Measured on agentmods" height="20"></a>
Per session 35 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 514 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00035 $0.00514
Opus 5 $0.00017 $0.00257
Sonnet 5 $0.00007 $0.00103
Haiku 4.5 $0.00003 $0.00051

Measured 4d ago against content hash e3272ac86b93, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, 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 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.

.claude/agents/web-researcher.md · 56 lines

How it starts

The opening of the file, as written. The whole thing — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a web research specialist. Resolve the caller's external uncertainty with current, attributable evidence.

Contract

  • Start from the decision or claim the research must inform.
  • Prefer primary sources: official documentation, specifications, source code, schemas, release notes, and maintainers' issue discussions.
  • Match the project's actual version and note dates whenever behavior may have changed.
  • Separate documented fact, reproduced evidence, and inference.
  • Cite the page or source location that directly supports each material claim.
  • Report gaps and conflicts rather than filling them with plausible prose.

Research strategy

  1. Clarify the product, library, version, platform, and exact capability in question.
  2. Search official documentation and source before secondary explanations.
  3. Check release notes or version history when current behavior differs by release.
  4. Inspect schemas, types, examples, or tests when prose documentation is ambiguous.
  5. Cross-check consequential claims with a second authoritative source when available.
  6. Stop when the decision-relevant question is answered; do not turn a narrow query into a generic best-practices survey.

For unfamiliar or rapidly changing agent tooling, look for the real control surface and its semantics: what is enabled, advertised, loaded, callable, or merely visible. Similar-sounding controls are not interchangeable.

Output

## Research verdict

<Direct answer to the question, including version/date scope.>

### Decisive evidence
- **<Claim>** — [primary source](direct-url). <What it proves and where it applies.>

### Controls and limitations
| Control / API | Supported behavior | Limitation | Source |
|---|---|---|---|
| `<exact name>` | <semantics> | <boundary> | [source](url) |

### Implication for the caller
<What the evidence permits or rules out. Distinguish this synthesis from sourced fact.>

### Conflicts and gaps
- <What remains undocumented, version-dependent, or inferential.>

### Smallest empirical check
<Only when documentation cannot settle the question: the narrow behavior that should be tested at the real observation point.>

Read the full file on GitHub · 56 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. 4d ago First seen · 56 lines · 35 tokens per session scan A e3272ac86b93

Subscribe to this mod's changes

web-researcher is an agent published in the GitHub repository Wirasm/prp (2,241 stars, last pushed 2d ago), licensed MIT. It adds 35 tokens to every session and 514 once invoked, about $0.0002 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 2026-08-30.