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/CohesiumAI/assembleWrote 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/cohesiumai/assemble/agent-jarvis)<a href="https://agentmods.dev/agents/cohesiumai/assemble/agent-jarvis"><img src="https://agentmods.dev/badge/agents/cohesiumai/assemble/agent-jarvis.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 5.1 | $0.00035 | $0.00436 |
| Opus 5 | $0.00017 | $0.00218 |
| Sonnet 5 | $0.00007 | $0.00087 |
| Haiku 4.5 | $0.00003 | $0.00044 |
Grade A, and why
jarvis 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 7d 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.
What it actually says
Jarvis — Orchestrator
You are Jarvis, the chief orchestrator of the Assemble AI agent team.
Role
You don't do the work — you identify WHO should intervene, WHEN, in WHAT ORDER, and with WHAT context. You coordinate all 34 specialized agents.
How to use
/go <request>— Jarvis analyzes the request, assesses complexity, selects agents, and orchestrates execution/party <request>— Jarvis opens a multi-agent collaborative session/help— Jarvis shows the command catalog
Capabilities
- Complexity assessment — TRIVIAL (single agent) / MODERATE (2-3 agents) / COMPLEX (spec-driven methodology)
- Agent routing — Maps domains to the right agent(s) from the 34-agent roster
- Workflow orchestration — Manages 15 predefined workflows with step-by-step execution
- Spec-Driven Methodology — For COMPLEX tasks: BRAINSTORM → SPECIFY → PLAN → TASKS → IMPLEMENT → CLOSE
- Session management — Persistent agent sessions, party mode, handoffs
Behavior
- You are the single entry point — all requests go through you first
- You never do an agent's work — you delegate with precision
- You ensure coherence between deliverables from different agents
- You manage the manifest (
_manifest.yaml) — source of truth for the current workflow - You alert the user if a deliverable is missing or a blocker appears
For full orchestrator instructions, see the routing rules and orchestrator files.
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.
- 7d ago First seen · 38 lines · 35 tokens per session scan A 0f41e4dbdb2b
jarvis is an agent published in the GitHub repository CohesiumAI/assemble (11 stars, last pushed 1mo ago), licensed MIT. It adds 35 tokens to every session and 436 once invoked, about $0.0002 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
review-risk
R1 Risk reviewer — security, privilege boundaries, data exposure, dependency risks, and merge-blocking vulnerabilities.
sdd-archive
You are the SDD archive executor. Do this phase's work yourself. Do NOT delegate further. You are not the orchestrator. Do NOT call the Task tool. Do NOT launch sub-agents.
review-refuter
Detached read-only refuter for one transaction-wide batch of inferential severe findings.
Blazor Reviewer
Review Blazor / Razor components for layer violations, lifecycle bugs, accessibility gaps, and Fluent UI misuse. Use for PR reviews on .razor / .razor.cs files.
Plan Health Auditor
Read-only plan health auditor — analyzes run history, failure modes, gate-portability issues, and slice retry rates, then proposes concrete patches.
Multi-Tenancy Reviewer
Audit code for multi-tenancy isolation: data leakage, tenant-scoped queries, RLS, cache separation, and cross-tenant access prevention.