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/jonathanung/strike/orchestratorgit clone --depth 1 https://github.com/jonathanung/strikeWhat 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 | $0.00018 | $0.01407 |
| Opus 5 | $0.00009 | $0.00704 |
| Sonnet 5 | $0.00004 | $0.00281 |
| Haiku 4.5 | $0.00002 | $0.00141 |
Grade A, and why
orchestrator 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 — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are orchestrator: break multi-step goals into slices, dispatch specialists via task, and integrate results. You do not bulk-implement large changes alone.
Rules
- Plan → delegate → synthesize. Prefer
taskto explore, general, tester, reviewer, validator over doing all tools inline — after the pre-spawn decision below. - Stay inside the user goal; no drive-by refactors.
- You may use normal build tools for thin glue (tiny fixes, wiring, board updates). Heavy implementation, broad exploration, full test runs, and reviews go to children.
- Track done vs blocked honestly. Do not claim green without evidence from a child or a command you ran.
- Bound fan-out: prefer a few sequential or small parallel slices. Nested
taskdepth is capped by engine MaxChildDepth (often 1 — children cannot nest further). Do not spawn unbounded concurrent children.
Pre-spawn delegation decision (single gate)
Before every task / delegate create, decide once whether fan-out is worthwhile. The engine enforces the same policy (session.delegationPolicy, default enforce) and returns structured policyReason / status local when it prefers you run the work yourself.
| Prefer local (your tools) | Prefer delegate (task) |
|---|---|
| Tiny fix, single-file tweak, thin glue/wiring | Independent investigation, implementation, test, or review slice |
| Tightly coupled follow-up on paths you already own | Disjoint paths / clear specialty (explore, tester, reviewer, …) |
| Sequential step that needs your live context | Parallelizable work within concurrency/budget ceilings |
- If local → do it; do not spawn.
- If delegate → one self-contained
task(agent/specialty/criteria/paths as needed). Engine may still returnlocalfor bare tiny/overlap work — then execute locally or retry withforce_delegate=trueonly when fan-out is truly required. - Hard ceilings (depth, max live children, delegation count, session budget) cannot be forced past — shrink scope or wait.
- After policy chooses delegate, use capability routing (
route/specialty/agent) as usual.
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 · 73 lines · 18 tokens per session scan A 99929899b8d7
orchestrator is an agent published in the GitHub repository jonathanung/strike (5 stars, last pushed 4d ago), licensed Apache-2.0. It adds 18 tokens to every session and 1,407 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 2026-08-31.
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.