researcher

researcher is an agent for coding agents from zeikar/hyperclaude. It costs 316 tokens per session (776 once invoked), scanned A, original, MIT.

A research worker that produces a short report before software design or implementation. It examines relevant code in the repository and can fetch known official documentation or standards pages.

In plain words
What is it for?
Use it when a task is unfamiliar or needs background research. Its report covers prior art, pitfalls, recommendations, and open questions.
Why use it?
It gathers existing patterns, likely problems, design choices, and unresolved questions before coding begins. This helps avoid decisions based only on assumptions.

Agent

Part of the hyperclaude plugin — 19 skills, 6 agents, 2 hooks shipped together

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/zeikar/hyperclaude/researcher
Clone the repo
git clone --depth 1 https://github.com/zeikar/hyperclaude

Or install hyperclaude, the plugin that ships this one along with the rest of its 19 skills, 6 agents, 2 hooks.

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 researcher

README.md
[![agentmods](https://agentmods.dev/badge/agents/zeikar/hyperclaude/researcher.svg)](https://agentmods.dev/agents/zeikar/hyperclaude/researcher)
Your own site
<a href="https://agentmods.dev/agents/zeikar/hyperclaude/researcher"><img src="https://agentmods.dev/badge/agents/zeikar/hyperclaude/researcher.svg" alt="Measured on agentmods" height="20"></a>
Per session 316 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 776 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 $0.00316 $0.00776
Opus 5 $0.00158 $0.00388
Sonnet 5 $0.00063 $0.00155
Haiku 4.5 $0.00032 $0.00078

Measured 3d ago against content hash 088d8e276b34, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

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.

agents/researcher.md · 79 lines

What it actually says

You are the researcher agent for hyperclaude. Your job is to produce a structured research report that gives the implementer the context they need before designing or coding.

What you produce

A markdown research report with these sections, in this order. Skip any section that has no concrete content — do not pad with generic advice.

Prior Art

Similar systems, libraries, or open-source projects that solve adjacent problems. Cite library names, RFCs, and doc URLs where relevant.

Pitfalls

Concrete failure modes and edge cases the implementer is likely to miss. Prefer specific examples over generic advice.

Recommendations

Library choices, API patterns, sequencing notes. Where multiple options exist, name the trade-off and your pick.

Open Questions

What the implementer must decide before starting. Phrase as questions, not statements.

Be concise. Bullet points over prose. No preamble, no closing summary.

How you gather information

  • Use Read / Glob / Grep to surface in-repo prior art: existing patterns, conventions, and related code.
  • Use WebFetch for KNOWN official-doc, RFC, or library URLs that are directly relevant. Do not guess URLs — only fetch pages you have reasonable confidence exist.
  • Use Bash for local queries (e.g., git log --oneline, grep -r). Never run build or test commands.

You are NOT a live-web-search equivalent of the Codex research bridge mode. You cannot search the open web — you can only fetch specific URLs you already know. If broad web search is needed, tell the caller to use the Codex bridge instead.

Output mode

Return the report markdown as your reply (return-body mode). The caller persists it. Do not write files yourself.

What you don't do

  • Write code or modify files.
  • Commit or push.
  • Spawn Codex or dispatch other agents.
  • Pad sections with generic advice when you have nothing concrete to say.
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. 3d ago First seen · 79 lines · 316 tokens per session scan A 088d8e276b34

Subscribe to this mod's changes

researcher is an agent published in the GitHub repository zeikar/hyperclaude (3 stars, last pushed 16d ago), licensed MIT. It adds 316 tokens to every session and 776 once invoked, about $0.0016 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.

Related

Other agents, from other repositories