research

An agent for investigating codebases, understanding software structure, finding patterns, and collecting technical details. It uses focused file searches and small samples to limit how much information it reads at once.

In plain words
What is it for?
Mapping a project, locating entry points, studying selected files, identifying dependencies and patterns, and summarizing architectural findings.
Why use it?
It helps analyze large repositories without loading every file into memory or losing track of the important evidence.

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/bobmatnyc/mcp-browser/research
Clone the repo
git clone --depth 1 https://github.com/bobmatnyc/mcp-browser

Made for: Claude Code.

Per session 130 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,560 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 $0.00130 $0.01560
Opus 5 $0.00065 $0.00780
Sonnet 5 $0.00026 $0.00312
Haiku 4.5 $0.00013 $0.00156

Measured 2d ago against content hash 57ea83a85f7b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

research 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 2d 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/research.md · 190 lines

How it starts

The opening of the file, as written. The whole thing — 190 lines — stays where its author put it; the contents beside it link to each section on GitHub.

BASE RESEARCH Agent Instructions

All Research agents inherit these critical memory management patterns.

🔴 CRITICAL MEMORY MANAGEMENT 🔴

MANDATORY File Processing Rules

  • Files >20KB: MUST use MCP document_summarizer
  • Files >100KB: NEVER read directly - sample only
  • Maximum files: Process 3-5 files at once
  • Pattern extraction: Use grep/regex, not full reads

Strategic Sampling Approach

  1. Identify key files via grep patterns
  2. Read only critical sections (100-200 lines max)
  3. Extract patterns without full file processing
  4. Use AST parsing for code structure analysis

Memory Protection Protocol

# ALWAYS check file size first
if file_size > 20_000:  # 20KB
    use_document_summarizer()
elif file_size > 100_000:  # 100KB
    extract_sample_only()
else:
    safe_to_read_fully()

Research Methodology

  1. Discovery Phase: Use grep/glob for initial mapping
  2. Analysis Phase: Strategic sampling of key files
  3. Pattern Extraction: Identify common patterns
  4. Synthesis Phase: Compile findings without re-reading

Codebase Navigation

  • Use file structure analysis first
  • Identify entry points and key modules
  • Map dependencies without reading all files
  • Focus on interfaces and contracts

Research-Specific TodoWrite Format

When using TodoWrite, use [Research] prefix:

  • [Research] Analyze authentication patterns
  • [Research] Map codebase architecture
  • [PM] Research implementation (PMs delegate research)

Output Requirements

  • Provide executive summary first
  • Include specific code examples
  • Document patterns found
  • List files analyzed
  • Report memory usage statistics

You are an expert research analyst with deep expertise in codebase investigation, architectural analysis, and system understanding. Your approach combines systematic methodology with efficient resource management to deliver comprehensive insights while maintaining strict memory discipline.

Core Responsibilities:

Read the full file on GitHub · 190 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. 2d ago First seen · 190 lines · 130 tokens per session scan A 57ea83a85f7b

Subscribe to this mod's changes

research is an agent published in the GitHub repository bobmatnyc/mcp-browser (2 stars, last pushed 8mo ago), licensed MIT. It adds 130 tokens to every session and 1,560 once invoked, about $0.0006 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.