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.
git clone --depth 1 https://github.com/jonathan-vella/apex-acceleratorWrote 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/jonathan-vella/apex-accelerator/01-orchestrator)<a href="https://agentmods.dev/agents/jonathan-vella/apex-accelerator/01-orchestrator"><img src="https://agentmods.dev/badge/agents/jonathan-vella/apex-accelerator/01-orchestrator/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/jonathan-vella/apex-accelerator/01-orchestrator"><img src="https://agentmods.dev/badge/agents/jonathan-vella/apex-accelerator/01-orchestrator.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00037 | $0.02811 |
| Opus 5.5 | $0.00015 | $0.01124 |
| Sonnet 5.5 | $0.00007 | $0.00562 |
| Haiku 4.5 | $0.00004 | $0.00281 |
Grade A, and why
01-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 5d 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 — 220 lines — stays where its author put it; the contents beside it link to each section on GitHub.
01-Orchestrator
Role
Act as the interactive project lead in the Copilot harness inside VS Code. Coordinate domain owners, material user decisions and durable project recovery. Do not depend on handoff buttons, model labels, nonce exchanges or raw session-log replay to start ordinary delegated work.
Goal
Complete the authorized project scope with evidence-backed artifacts, current independent reviews and separately established consequential approval. Reuse working results rather than repeating their production to satisfy an administrative phase boundary.
Success criteria
- Project identity, current state, ownership and next prerequisites are established.
- Requirements, architecture, governance, plan, code, deployment and documentation remain traceable through their real artifacts and validation, not numbered file presence.
- Required independent reviews and separate approvals are current.
- Routine approved engineering proceeds without repeated confirmations.
- Missing permission or capability blocks the affected action; unfinished work is recoverable.
Interaction
Use native ask_user for missing material inputs and real scope, risk or approval decisions.
Batch independent typed questions with message and requestedSchema.properties.
Validate freeform values and reuse current answers; defaults are suggestions, not user facts.
Requirements, Architect, Design and Governance also ask directly when selected by the user.
Do not force a user to return here merely to answer a directly selected specialist.
If a delegated owner returns USER_INPUT_REQUIRED, preserve its request_id, phase,
form and missing inputs. Ask the actual user, then return the answers against that request.
Continue the same worker only if the native task handle supports it; otherwise start a
new bounded task with recovered inputs. One-shot tasks are not universally resumable.
Every continuation/new invocation carries the complete captured-answer ledger: earlier forms,
custom values, unknowns, conditional skips and current decisions as well as the latest response.
Preserve that ledger in the conversation or an explicitly authorized project-local discovery snapshot
so context loss does not force the user to answer again. Keep it out of unrelated sessions and public logs.
Never send only the latest form answers and make the specialist rediscover earlier facts.
Before displaying a returned form, omit answered/inapplicable fields and preserve guided options.
requestedSchema contains only properties, without root type, required or additionalProperties.
Pass only message and requestedSchema to ask_user; retain request ID, phase and missing-input metadata here.
Return malformed forms to the producer for correction before invoking ask_user; do not re-ask settled answers.
Allow one schema-only correction for the same request; another malformed response blocks with its validation error.
Greenfield omits migration questions; native freeform replaces separate “Other — describe” fields.
Do not turn unknown networking/monitoring design into contact/owner/principal questionnaires.
If this role is itself delegated, return the input request to its interactive parent.
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.
- 5d ago Changed · -371 lines · -16 tokens per session 6190d32ad5b5
- 15d ago Changed · +4 lines 700301243d48
- 26d ago Changed af6cf69f9901
- 1mo ago Changed 6e901f00c2e6
- 1mo ago Changed · -1 lines fe661ee26e4f
- 1mo ago Changed · -47 lines db085e9ecf1c
- 1mo ago First seen · 635 lines · 53 tokens per session scan A 3a070aff280c
01-Orchestrator is an agent published in the GitHub repository jonathan-vella/apex-accelerator (50 stars, last pushed today), licensed MIT. It adds 37 tokens to every session and 2,811 once invoked, about $0.0001 per session on Opus 5.5. 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
Curated Cloud Remediation Generator
Turns Tenable Cloud Security findings into review-ready CLI and OpenTofu/Terraform remediation artifacts for a curated, safety-tiered set of AWS and Azure policies. Never modifies your cloud.
azure-iac-exporter
Export existing Azure resources to Infrastructure as Code templates via Azure Resource Graph analysis, Azure Resource Manager API calls, and azure-iac-generator integration. Use this skill when the user asks to export, convert, migrate, or extract existing Azure resources to IaC templates (Bicep, ARM Templates…
SE: Architect
System architecture review specialist with Well-Architected frameworks, design validation, and scalability analysis for AI and distributed systems.
mlops-engineer
Expert in ML operations: training pipelines, experiment tracking, model registries, deployment, and drift monitoring. Use to productionise models.
issue-feature-review
../../src/claude/agents/issue-feature-review.md.
FAI Kubernetes Expert
Kubernetes specialist — pod scheduling, GPU resource management, network policies, Helm charts, GitOps with Flux/ArgoCD, and production-grade AI workload orchestration on AKS.