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 mateo-cuello/orchestrator-claude--agy --skill agy-orchestrationgit clone --depth 1 https://github.com/mateo-cuello/orchestrator-claude--agyWrote 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/mateo-cuello/orchestrator-claude--agy/agy-orchestration)<a href="https://agentmods.dev/skills/mateo-cuello/orchestrator-claude--agy/agy-orchestration"><img src="https://agentmods.dev/badge/skills/mateo-cuello/orchestrator-claude--agy/agy-orchestration/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/mateo-cuello/orchestrator-claude--agy/agy-orchestration"><img src="https://agentmods.dev/badge/skills/mateo-cuello/orchestrator-claude--agy/agy-orchestration.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 | $0.00065 | $0.02672 |
| Opus 5 | $0.00032 | $0.01336 |
| Sonnet 5 | $0.00013 | $0.00534 |
| Haiku 4.5 | $0.00006 | $0.00267 |
Grade B, and why
agy-orchestration scanned grade B with 1 finding 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 11d 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.
Unrestricted tool accessmediumExcessive agency
A wildcard tool grant or "run any command" leaves no least-privilege boundary at all.
Workers **cannot** run arbitrary commands (`command(*)` is set to `ask` and auto-denies in How it starts
The opening of the file, as written. The whole thing — 196 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Orchestrating agy (Gemini) sub-agents
On install: copy this directory to
~/.claude/skills/agy-orchestration/and replaceC:\path\to\clagywith the real path where you cloned the repo. The paths must be absolute and real so they can be read at runtime.
Tools served by the bridge at C:\path\to\clagy\mcp\agy-bridge.mjs:
| blocking | in the background | |
|---|---|---|
agy_dispatch — one task |
agy_start — returns a jobId instantly |
|
agy_fanout — N in parallel |
agy_result / agy_jobs — status and result |
The blocking ones are the default. Use agy_start when there's something else to do
meanwhile: the Stop hook injects the result when the turn closes, no polling needed. Careful
with that last point — the notice lands on a turn boundary, so launching a job and closing
right away buys nothing over agy_dispatch. And if the next step depends on the result,
blocking is simpler.
No subagent in the middle
There is no agy-researcher or agy-auditor to hand this to, and reaching for a general
Claude subagent instead is the one move that defeats the purpose. A subagent's loop is always
Claude: delegating to one spends the resource the workers exist to conserve, and it spends it
before any Gemini token is bought. The main thread composes the fan-out and calls the tools
itself.
If a capability can't be delegated to a worker, drop it — never substitute a Claude subagent. The measured failure: one research task, ~162k Claude tokens,
agy_jobsempty.
So the composing is mine: split the question, scope each cwd, pick modes and models. So is
the checking — and it's cheap, because I already know which line to look at. Compare every
cited file against trace.filesTouched, which the bridge takes from the worker's step stream
rather than its self-report: a finding citing a file that isn't in there is invented.
confidence: 1 is not evidence; these workers report full confidence with the tests red.
Detail: C:\path\to\clagy\docs\native-agents.md.
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.
- 11d ago First seen · 196 lines · 65 tokens per session scan B 33c2ba9cb91c
agy-orchestration is a skill published in the GitHub repository mateo-cuello/orchestrator-claude--agy (2 stars, last pushed 1mo ago), licensed MIT. It adds 65 tokens to every session and 2,672 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it B with 1 finding (unrestricted tool access). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
agent-self-scheduling
Schedule AI agent runs with cron, loops, or external clocks while avoiding unsafe tight autonomous timers.
cao-plugin
Create a new CAO (CLI Agent Orchestrator) plugin. Use this skill whenever the user wants to add a plugin that reacts to CAO lifecycle or messaging events, scaffold a plugin package, understand plugin requirements, or integrate an external system (Discord, Slack, dashboards, logging, metrics) with CAO. Also use when…
cao-session-management
Interact with CAO (CLI Agent Orchestrator) — launch multi-agent sessions, check status, send follow-up instructions, unblock stuck terminals, or shut down sessions. Use when working with CAO sessions in any capacity.
assimilate-popular-workflows
This skill should be used when the user asks to "find skills in the wild", "assimilate popular workflows", "discover SKILL.md files in repos", "research external skills", "find workflow patterns", "survey the skill landscape", "what skills exist out there", or wants to investigate public repositories for extractable…
mcp-app-verification
Comprehensive verification checklists for MCP Apps. Tests with basic-host reference, validates handler-before-connect, text fallback, resource URI linking, single-file bundling, host styling, CSP, and legacy pattern detection.
mcp-host-styling-integration
Integrates MCP App UI with host theming system. Applies host CSS variables, handles onhostcontextchanged, safe area insets, display mode detection, and fullscreen configuration.