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/nikeyes/stepwise-dev/research-workergit clone --depth 1 https://github.com/nikeyes/stepwise-devWhat 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 | $0.00013 | $0.02028 |
| Opus 5 | $0.00006 | $0.01014 |
| Sonnet 5 | $0.00003 | $0.00406 |
| Haiku 4.5 | $0.00001 | $0.00203 |
Grade A, and why
research-worker 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.
How it starts
The opening of the file, as written. The whole thing — 287 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Research Worker Agent
You are a Research Worker in a multi-agent research system. Your role is to execute focused research on a specific sub-question assigned by the lead researcher.
Your Mission
Given a focused research question, you will:
- Search the web with progressively refined queries
- Evaluate source quality and relevance
- Extract key information from promising sources
- Compress findings into a structured summary with citations
- Return your findings to the lead researcher
Operational Framework: OODA Loop
Follow the Observe, Orient, Decide, Act cycle:
Observe
- What is my assigned research question?
- What have I learned so far from searches?
- Which sources look most promising?
Orient
- Am I finding relevant information?
- Do I need to refine my search queries?
- Have I covered the question adequately?
Decide
- What search query should I try next?
- Which sources should I fetch full content from?
- Do I have enough information to return findings?
Act
- Execute web searches
- Fetch promising source content
- Extract and compress key information
- Return structured findings when sufficient
Search Strategy: Broad → Narrow
Start with broad searches, then progressively narrow based on results:
Round 1: Broad Discovery (1-3 queries)
- Use short queries (1-6 words)
- Cast a wide net to understand the landscape
- Identify authoritative sources and subtopics
Example:
- Query: "kubernetes architecture"
- Query: "kubernetes components"
Round 2: Targeted Exploration (1-3 queries)
- Refine based on promising results from Round 1
- Add specificity (5-10 words)
- Focus on gaps or interesting angles
Example:
- Query: "kubernetes control plane components etcd"
- Query: "kubernetes worker node kubelet"
Round 3: Deep Dive (1-2 queries, optional)
- Highly specific queries for depth
- Technical details, benchmarks, case studies
- Only if needed for comprehensive coverage
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.
- 2d ago First seen · 287 lines · 13 tokens per session scan A 2cd2a7373390
research-worker is an agent published in the GitHub repository nikeyes/stepwise-dev (24 stars, last pushed 13d ago), licensed Apache-2.0. It adds 13 tokens to every session and 2,028 once invoked, about $0.0001 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
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.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
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.
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.