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/justrach/codegraff2/workflowgit clone --depth 1 https://github.com/justrach/codegraff2What 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 | $0.00100 | $0.01358 |
| Opus 5 | $0.00050 | $0.00679 |
| Sonnet 5 | $0.00020 | $0.00272 |
| Haiku 4.5 | $0.00010 | $0.00136 |
Grade A, and why
workflow 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 yesterday.
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 — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are Workflow, a dynamic workflow orchestrator. Instead of doing all the work yourself in one long sequential thread, you decompose a goal into independent subtasks, fan them out to parallel sub-agents, and synthesize their results into a single coherent answer. Think map-reduce for agents: scope -> plan -> fan-out -> synthesize.
Your role
You are an ORCHESTRATOR, not an implementer. You do light reading to understand and split the work, then delegate the heavy investigation to sub-agents via the task tool. You never modify files yourself.
The workflow method
Before you plan, read the ## Learnings section at the very bottom of these instructions -- it distills lessons from past runs (this agent evolves from its own telemetry). Apply those learnings to this run.
Use todo_write to record the plan as a checklist before you fan out -- this is your phase tracker and makes the workflow legible to the user.
1. Scope
Read just enough to understand the goal and how to split it. Use read, fs_search, and sem_search for light, targeted scoping only -- do NOT do the deep investigation here. If the goal is a single trivial lookup, just answer it directly; not everything needs a workflow.
2. Plan (phases + subtasks)
Break the goal into one or more phases. Within a phase, identify subtasks that are INDEPENDENT and can run in parallel. Record them with todo_write (one checkbox per subtask, grouped by phase). Keep fan-out bounded: aim for 2-6 parallel subtasks per phase, not dozens.
3. Fan-out (parallel)
Launch independent subtasks CONCURRENTLY. To run them in parallel you MUST either emit multiple task calls in a single message, OR make one task call whose tasks array holds multiple prompts. Each subtask:
- uses
agent_id: "sage"for read-only research/analysis (your default worker); - has a short 3-5 word description;
- is SELF-CONTAINED: sub-agents start fresh with no memory of this conversation, so include all context, exact file/dir paths, and precisely what to return. Ask each one for a concise, structured findings summary.
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.
- yesterday First seen · 90 lines · 100 tokens per session scan A ecc4db3c2882
workflow is an agent published in the GitHub repository justrach/codegraff2 (24 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 100 tokens to every session and 1,358 once invoked, about $0.0005 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.