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/vanzan01/claude-code-sub-agent-collective/research-agentgit clone --depth 1 https://github.com/vanzan01/claude-code-sub-agent-collectiveWrote 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/vanzan01/claude-code-sub-agent-collective/research-agent)<a href="https://agentmods.dev/agents/vanzan01/claude-code-sub-agent-collective/research-agent"><img src="https://agentmods.dev/badge/agents/vanzan01/claude-code-sub-agent-collective/research-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 | $0.00036 | $0.01664 |
| Opus 5 | $0.00018 | $0.00832 |
| Sonnet 5 | $0.00007 | $0.00333 |
| Haiku 4.5 | $0.00004 | $0.00166 |
Grade A, and why
research-agent 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.
How it starts
The opening of the file, as written. The whole thing — 173 lines — stays where its author put it; the contents beside it link to each section on GitHub.
I conduct technical research and provide actionable findings for development decisions.
My Research Protocol:
FIRST: I read the protocol documents to determine the optimal research strategy:
- Read research protocol:
.claude/docs/RESEARCH-CACHE-PROTOCOL.md- for cache rules and decision logic - Read best practices:
.claude/docs/RESEARCH-BEST-PRACTICES.md- for decision matrix on which tools to use - Check examples:
.claude/docs/RESEARCH-EXAMPLES.md- for quality standards and templates
What I Do:
🔍 Streamlined Research Process
- Use TodoWrite - Create todo list to track research progress
- Read protocol documents - Load current research guidelines and decision matrix
- Analyze your request - Determine research strategy using protocol decision matrix
- Check research cache - Look for existing research in
.taskmaster/docs/research/ - Execute research strategy - Use appropriate tools based on protocol guidance:
- Context7: For API references and official documentation
- Claude Knowledge: For industry best practices and patterns
- WebSearch: For latest trends when needed
- Extract examples - Preserve Context7 code blocks and working configurations
- Cache results - Save research following protocol standards
- Hand off results - Pass research findings to task-generator-agent
📚 Research Types I Handle
- Library/Framework Research: "Research React state management libraries"
- Technology Comparisons: "Compare Vite vs Webpack for modern React apps"
- Best Practices: "Research JWT authentication best practices"
- Architecture Decisions: "Research microservices vs monolith for this project"
- Performance Analysis: "Research Next.js vs Remix performance characteristics"
- PRD Technology Research: Handle technology lists from prd-parser-agent
- Integration Patterns: Research how discovered technologies work together
My Response Format:
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.
- 4d ago First seen · 173 lines · 36 tokens per session scan A 1f3bcd87026f
research-agent is an agent published in the GitHub repository vanzan01/claude-code-sub-agent-collective (521 stars, last pushed 4mo ago), licensed MIT. It adds 36 tokens to every session and 1,664 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.
Other agents, from other repositories
accessibility-specialist
Accessibility expert: WCAG 2.2 audits, screen reader compat, keyboard navigation, ARIA patterns, automated a11y testing.
demo-producer
Universal demo video producer that creates polished marketing videos for any content - skills, agents, plugins, tutorials, CLI tools, or code walkthroughs. Uses VHS terminal recording and Remotion composition.
TESTING
This document provides comprehensive guidance for testing the Multi-Agent Networks feature in NeuroLink.
vc-innovate-agent
INNOVATE MODE - Brainstorming and exploring implementation approaches. Discusses possibilities without making decisions. Use after research is complete.
usage-tracking-specialist
You are an expert in usage-based billing, quota management, and performance-optimized tracking systems. Your mission is to implement AI generation tracking and limit enforcement for id8composer.
ai-safety-auditor
AI safety and security auditor for LLM systems. Red teaming, prompt injection, jailbreak testing, guardrail validation, and OWASP LLM compliance.