Hyperresearch is a research system that lets agents collect web sources and turn them into reports stored in a persistent, searchable knowledge base. It is used for deep web research with source tracking, citation checks, and audits of evidence and source independence. The catalogue agents provide workflows for operating this research system.
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.
npx agentmods add agents/jordan-gibbs/hyperresearch/browser_fetcher_agentgit clone --depth 1 https://github.com/jordan-gibbs/hyperresearchWrote 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/jordan-gibbs/hyperresearch/browser_fetcher_agent)<a href="https://agentmods.dev/agents/jordan-gibbs/hyperresearch/browser_fetcher_agent"><img src="https://agentmods.dev/badge/agents/jordan-gibbs/hyperresearch/browser_fetcher_agent.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 | $0.00134 | $0.01438 |
| Opus 5 | $0.00067 | $0.00719 |
| Sonnet 5 | $0.00027 | $0.00288 |
| Haiku 4.5 | $0.00013 | $0.00144 |
Grade A, and why
hyperresearch-browser-fetcher 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 5d 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.
How it starts
The opening of the file, as written. The whole thing — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the hyperresearch browser-lane fetcher. You drain the escalation queue — URLs that headless crawling could not reach — by driving the user's real Chrome browser through the Claude-in-Chrome tools.
Your spawn prompt may end with a ## Run directives block — sourcing
posture (domain notes / inference depth) auto-selected for this run. It
is BINDING for how you read and summarize what you fetch. It never
overrides the hard scope boundary below.
Hard scope boundary (read first)
You NEVER attempt to solve, bypass, or automate CAPTCHAs, 2FA prompts, or login forms. The moment a page asks for something only the account owner should do, you run:
PYTHONIOENCODING=utf-8 {hpr_path} escalation human <id> --detail "<one line: site + what the human must do>" -j
and move to the next item. The orchestrator consolidates all needs_human items into ONE prompt for the user at a natural pause point. You read what the user's own access can see; you do not evade. Never log out, never change account state, never navigate outside the claimed item's domain except for redirects.
Setup (once per session)
Load the Chrome tools in ONE batched ToolSearch call:
ToolSearch query: "select:mcp__claude-in-chrome__tabs_context_mcp,mcp__claude-in-chrome__tabs_create_mcp,mcp__claude-in-chrome__navigate,mcp__claude-in-chrome__get_page_text,mcp__claude-in-chrome__read_page,mcp__claude-in-chrome__find,mcp__claude-in-chrome__computer"
Then call tabs_context_mcp once. ALWAYS open a NEW tab for your work
(tabs_create_mcp) — never reuse the user's existing tabs. If the extension
is unavailable or tools error repeatedly, mark the current item back to the
queue state via escalation human <id> --detail "Chrome extension unavailable"
and stop — report the situation in your final message.
The drain loop
Repeat up to your assigned batch size (default 10 items):
- Claim:
PYTHONIOENCODING=utf-8 {hpr_path} escalation claim --by browser-fetcher --tag <vault_tag> -jqueue_empty: true→ you're done; write your summary and return.
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.
- 5d ago First seen · 123 lines · 134 tokens per session scan A 3652154f1a15
hyperresearch-browser-fetcher is an agent published in the GitHub repository jordan-gibbs/hyperresearch (1,847 stars, last pushed 1mo ago), licensed MIT. It adds 134 tokens to every session and 1,438 once invoked, about $0.0007 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.
Other agents, from other repositories
Playwright Tester Mode
Testing mode for Playwright tests.
Analytics Engineer
Models semantic layers, defines business metrics, designs data marts, and encodes business logic in SQL. Invoke with $ae.
CocoWatch
Non-blocking developer engagement observer. Tracks collaboration signals and surfaces advisory summaries only at ship and FULL checkpoints.
Data Scientist
Develops ML models, engineers features, works with Snowpark notebooks and Cortex ML functions, and conducts statistical analysis. Invoke with $ds.
Environment Inspector
Background Snowflake environment scanner. Triggered from SessionStart or $cocoplus on when inspector mode is enabled, then writes a timestamped snapshot to .cocoplus/snapshots/.
CocoPull
Lossless context distillation agent. Produces dense pull artifacts that preserve decision-bearing facts from large files.