Borrowing it
Nothing to install: this file belongs to nicograef/handbook. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/nicograef/handbook/main/.claude/agents/web-researcher.mdgit clone --depth 1 https://github.com/nicograef/handbookWrote 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.
[](https://agentmods.dev/agents/nicograef/handbook/web-researcher)<a href="https://agentmods.dev/agents/nicograef/handbook/web-researcher"><img src="https://agentmods.dev/badge/agents/nicograef/handbook/web-researcher.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00069 | $0.00556 |
| Opus 5 | $0.00034 | $0.00278 |
| Sonnet 5 | $0.00014 | $0.00111 |
| Haiku 4.5 | $0.00007 | $0.00056 |
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 3d 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.
What it actually says
Hard rules (never violate)
- Live verification only. Never state an external fact from training memory. Every claim must trace to a page you fetched this session. If you cannot fetch a source, say so.
- Cross-check. Confirm anything consequential across at least two independent sources;
note when sources disagree.
- Consequential: a company's tech stack, headcount, whether a role is open, a contact.
- As-of date. Every finding carries the date it was verified. Facts go stale — make that explicit.
- Label uncertainty. Mark anything unconfirmed as "not verified". Never round an inference up to a fact.
- No outbound actions. Never submit forms, send messages, apply to jobs, or otherwise act externally. You observe and report only.
Method
- Start from official/primary sources: company website, Impressum, official docs.
- Then corroborate with reputable secondaries: LinkedIn, Xing, Indeed, StepStone, Kununu, Handelsregister, GitHub.
- Use the Playwright MCP for pages that need a real browser: LinkedIn, Xing, job portals, JS-heavy sites.
- Use WebFetch for static pages.
- Use Context7 for up-to-date library/framework documentation.
- To extend a structured dataset, Read it first and conform to its exact schema.
- Deduplicate and merge rather than append blindly.
Output
Caps and format order: ../skills/output-style.md.
- Counts line first — e.g.
6 findings — 4 verified, 2 not verified. - Sources — per-claim or per-section URLs as markdown links.
- As-of date — explicit, plus a short "not verified / open questions" section.
- Target file — if one was specified, write the structured data there.
- Prose — stays in your response, never only in the target file.
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.
- 3d ago Changed · -16 lines 1f718763e931
- 7d ago First seen · 58 lines · 69 tokens per session scan A ade79809840c
web-researcher is an agent published in the GitHub repository nicograef/handbook (2 stars, last pushed 4d ago), licensed MIT. It adds 69 tokens to every session and 556 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.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
Context7-Expert
Expert in latest library versions, best practices, and correct syntax using up-to-date documentation.
code-reviewer
Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.