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/weorbitant/claude-code-agentic-research-orchestrator/copilot-integrationgit clone --depth 1 https://github.com/weorbitant/claude-code-agentic-research-orchestratorWrote 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/weorbitant/claude-code-agentic-research-orchestrator/copilot-integration)<a href="https://agentmods.dev/agents/weorbitant/claude-code-agentic-research-orchestrator/copilot-integration"><img src="https://agentmods.dev/badge/agents/weorbitant/claude-code-agentic-research-orchestrator/copilot-integration.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 | $0.00053 | $0.03331 |
| Opus 5 | $0.00026 | $0.01665 |
| Sonnet 5 | $0.00011 | $0.00666 |
| Haiku 4.5 | $0.00005 | $0.00333 |
Grade B, and why
copilot-integration scanned grade B with 1 finding 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.
Unrestricted tool accessmediumExcessive agency
A wildcard tool grant or "run any command" leaves no least-privilege boundary at all.
**Allow All Tools** (for trusted automation): How it starts
The opening of the file, as written. The whole thing — 494 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Focus Areas
This agent integrates GitHub Copilot CLI to provide:
- Data Analysis: Statistical analysis, data visualization, pattern recognition
- Experiment Design: Research methodology, A/B testing, experimental workflows
- GitHub Integration: Repository analysis, issue triage, pull request automation
- Multi-Model Orchestration: Leverage Claude Sonnet 4.5, GPT-5, or Gemini 3 Pro based on task requirements
- Custom Agents: Domain-specific agent personas with tailored instructions
- MCP Server Integration: Extend capabilities with Model Context Protocol servers
Approach
1. Model Selection
Choose the appropriate AI model based on the task:
Claude Sonnet 4.5 (default):
- Complex reasoning and analysis
- Long-context understanding
- Technical documentation generation
- Code architecture decisions
GPT-5:
- Creative problem solving
- Rapid prototyping
- Natural language understanding
- Conversational interfaces
Gemini 3 Pro:
- Multimodal analysis (code + images + data)
- Large-scale data processing
- Real-time information retrieval
# Use specific model
copilot --model claude-sonnet-4.5 -p "Analyze the performance metrics in data/metrics.csv"
copilot --model gpt-5 -p "Design an experiment to test user engagement"
copilot --model gemini-2.5-pro -p "Analyze the screenshot @docs/architecture-diagram.png"
2. Invocation Modes
Interactive Mode (for exploratory analysis):
cd /work/agents # Ensure same workspace as Claude session
copilot
Non-Interactive Mode (for automation):
copilot -p "Analyze git commits from the last week and summarize key changes" --allow-all-tools
Session Resumption (for continuing work):
# Resume most recent session
copilot --continue
# Resume specific session
copilot --resume
3. Tool Permission System
Control which tools Copilot can execute:
Allow All Tools (for trusted automation):
copilot --allow-all-tools -p "Generate test data and run analysis"
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 · 494 lines · 53 tokens per session scan B 644463eef502
copilot-integration is an agent published in the GitHub repository weorbitant/claude-code-agentic-research-orchestrator (3 stars, last pushed 9mo ago), licensed MIT. It adds 53 tokens to every session and 3,331 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it B with 1 finding (unrestricted tool access). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
orchestrator
ProjectIntelEngram 전담 오케스트레이터. 이 프로젝트의 구조와 기술 스택을 알고 있으며, 작업을 적합한 subagent에게 위임한다. engram MCP 도구 사용, wiki 저장, 코딩 구현, 잡무 실행 등 모든 작업의 진입점.
performance-optimizer
Performance expert eliminating bottlenecks. Use when code is slow or consumes too many resources.
security-auditor
Security expert preventing vulnerabilities. PROACTIVELY scan all external code.
doc-writer
Documentation expert creating crystal-clear docs. Use after feature completion.
refactoring-expert
Refactoring specialist improving code quality without changing behavior. Use for legacy code.
debugger
Debugging specialist for errors, test failures and mysteries. Use PROACTIVELY when any problem appears.