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 skills add Arvo-AI/aurora --skill mcpgit clone --depth 1 https://github.com/Arvo-AI/auroraWrote 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/skills/arvo-ai/aurora/mcp)<a href="https://agentmods.dev/skills/arvo-ai/aurora/mcp"><img src="https://agentmods.dev/badge/skills/arvo-ai/aurora/mcp/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/skills/arvo-ai/aurora/mcp"><img src="https://agentmods.dev/badge/skills/arvo-ai/aurora/mcp.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 | $0.00027 | $0.00640 |
| Opus 5.5 | $0.00011 | $0.00256 |
| Sonnet 5.5 | $0.00005 | $0.00128 |
| Haiku 4.5 | $0.00003 | $0.00064 |
Grade A, and why
mcp 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.
How it starts
The opening of the file, as written. The whole thing — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Custom MCP Servers
Overview
Your organisation has registered one or more of its own MCP (Model Context Protocol) servers. These expose systems Aurora has no built-in connector for, so they are often the ONLY way to see telemetry or configuration for those systems. Treat them as first-class investigation sources, not a last resort.
Their individual tools are NOT listed in your prompt — there can be hundreds. You reach
them through two tools: mcp_list_tools to discover, then mcp_call_tool to invoke.
Connected servers
{mcp_servers_section}
Instructions
- Narrow, do not dump.
mcp_list_tools()with no arguments gives the server list and counts.mcp_list_tools(query="...")searches across every server at once — prefer this when you do not know which server holds what you need, because it is far smaller than listing a whole server.mcp_list_tools(server="...")lists one server's tools. - Listings show
argsasname:type, with?meaning optional. That is usually enough to build the call. When a tool's arguments are complex or you need the exact schema, ask for the one tool:mcp_list_tools(server="...", tool="...")returns its full inputSchema. - Call with exact names. Pass the
serverandtoolstrings from discovery verbatim tomcp_call_tool. Guessing a name wastes a turn; the call is refused with the server list. A missing required argument comes back naming it, along with the schema. - Prefer a specific tool over a broad one. If a server offers both
get_zoneandlist_zones, fetch the single record you need rather than listing everything. - Report which server answered. When a finding comes from a custom MCP server, name the server label in your conclusion so a human can verify it against the right system.
- A failure is information. If a tool returns a connection error, say so plainly and move on to another source. Do not retry the same call repeatedly — the server may be unreachable from Aurora's network, which is a real finding worth reporting.
- Background RCA is allow-only. Confirm tools are omitted from discovery and cannot be called when no user is present to approve them — plan around allow tools only.
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 · 53 lines · 27 tokens per session scan A ab2c52d0fa82
mcp is a skill published in the GitHub repository Arvo-AI/aurora (434 stars, last pushed today), licensed Apache-2.0. It adds 27 tokens to every session and 640 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 2026-10-09.
Other skills, from other repositories
isaac-automator
Deploy and operate a cloud Isaac Workstation with Isaac Automator: provision a GPU VM running Isaac Sim, Isaac Lab, and/or Isaac Lab Arena on AWS, GCP, Azure, or Alibaba Cloud, connect to it, move data in and out, control cost with stop/start, repair, import existing deployments, and destroy. Use when the user wants a…
cloud-iam-deep
GCP/AWS/Azure cloud exploitation -- Cloud Functions, Firestore, Cloud Run, S3, MinIO, Blob Storage, SA keys.
performing-cloud-forensics-investigation
Conduct forensic investigations in cloud environments by collecting and analyzing logs, snapshots, and metadata from AWS, Azure, and GCP services.
databricks-platform-provisioning
Provision and test Databricks workspaces. Use when the user asks to create a workspace, set up a new environment, provision infrastructure, bootstrap Databricks, test a workspace, verify a deployment, or run validation checks against a Databricks workspace. Covers Azure, AWS, and GCP.
databricks-deployment-verification
MANDATORY post-deployment verification. After any workspace + Unity Catalog deployment, you MUST run all THREE compute paths against a UC table — classic cluster, serverless SQL warehouse, serverless notebook job. Skipping any of the three is incomplete work. Use whenever a workspace + UC has been freshly deployed or…
databricks-identity-governance
Manage Databricks identity and governance. Use when the user asks to create groups, users, service principals, set up RBAC, manage permissions, workspace assignments, or configure access control.