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 Maxmedawar/tandem --skill tandem-orchestrationgit clone --depth 1 https://github.com/Maxmedawar/tandemWrote 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/maxmedawar/tandem/tandem-orchestration)<a href="https://agentmods.dev/skills/maxmedawar/tandem/tandem-orchestration"><img src="https://agentmods.dev/badge/skills/maxmedawar/tandem/tandem-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/maxmedawar/tandem/tandem-orchestration"><img src="https://agentmods.dev/badge/skills/maxmedawar/tandem/tandem-orchestration.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
SkillSpector: 1 finding, up to low
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- low Excessive Agency · line 10 Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.Fix: Limit the skill's scope to its documented purpose. Remove instructions that enable the agent to perform actions outside its stated functionality.
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 | $0.00094 | $0.01624 |
| Opus 5 | $0.00047 | $0.00812 |
| Sonnet 5 | $0.00019 | $0.00325 |
| Haiku 4.5 | $0.00009 | $0.00162 |
Grade A, and why
tandem-orchestration 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 10d 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 — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Tandem orchestration
🏗️🏗️🏗️🏗️🏗️ Full orchestration is active.
Tandem lets an MCP-capable chat, coding agent, desktop client, or automation operate live agent and terminal sessions across the user's own machines. It is not tied to Claude.ai and it is not limited to Claude Code.
Supported engines in this MVP:
| Engine | What Tandem drives | Default |
|---|---|---|
claude |
Claude Code in tmux | enabled |
codex |
Codex in tmux | opt-in |
shell |
The user's login shell in tmux | opt-in |
hermes |
An existing allowlisted Hermes WebUI agent | opt-in |
The general session tools are engine-neutral. The built-in relay is intentionally Claude-only because its persistent manager protocol depends on Claude-specific behavior.
Safety rules
- Treat every session call as real execution on a real machine. Agents and shells retain the permissions of the OS account running Tandem.
- Never widen
TANDEM_CWD_ALLOWLISTmerely to get around a failure. It is an admission boundary, not a sandbox. - Never reveal tokens, paths, usernames, hostnames, tailnet identity, nonces, or raw fleet connection data in chat or reports.
- Use
shellonly when the user explicitly enabled it and the caller is trusted with arbitrary OS-user command execution. - Permission bypass is off by default. Never assume it is enabled. It applies only to Claude and is required only for the unattended Claude relay.
- Prefer a fresh named session for a new task or phase. Reuse a session only when continuity is valuable and its current state has been inspected.
- Never resend a prompt just because a turn is still running. Poll with an empty
textand the returned cursor. - Ask before destructive, irreversible, or materially broader actions. Tandem does not grant authority beyond the user's request.
Start every run with the fleet
Call list_devices first. It returns only:
{
"devices": [
{ "id": "local", "name": "local", "online": true, "engines": ["claude", "codex"] }
]
}
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.
- 10d ago First seen · 146 lines · 94 tokens per session scan A bb760a6b8795
tandem-orchestration is a skill published in the GitHub repository Maxmedawar/tandem (25 stars, last pushed 28d ago), licensed MIT. It adds 94 tokens to every session and 1,624 once invoked, about $0.0005 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 skills, from other repositories
adk-unit-design
Writes an as-built architecture document for one ADK code unit — purpose, execution flow, data flow, cross-class dependencies, extension points, and the parts that must not change — to docs/design/{topic}/{unit}/index.md. It describes the code as implemented, not a proposed design, and its reader is a developer about…
artifact-deploy
One-click deploy a user's pre-built app/artifact into their OWN AWS account and get a global public HTTPS link (Vercel-like), with a default TTL and promote-to-persistent. Use when the user says "deploy this", "ship this demo", "give me a public link", "share this externally", or "deploy to AWS".
explain-for
Explain a topic, a piece of code, an error, or a design decision calibrated to one named audience — a 5-year-old, a 5th grader, a manager, a designer, a graduate student, a parent. Resolves who the explanation is for (from the request, or from what memory already records about that person), establishes the ground…
web-verify
Look at your OWN front-end change before claiming it works -- navigate the loopback URL of a dev server or pod you started, screenshot the surface you changed, read the image to judge it, and embed it in chat. Three capture backends: playwright-cli (the session the dashboard Browser panel shows), the agent-browser CLI…
learn-from-sage
Detection-gap (miss) analysis for Code Review Sage. Learn from shipped fixes, acted-on human comments, and design outcomes to close reviewer blind spots. Inline during review stages a candidate; a human triggers a one-shot AI consolidation into the live ruleset.
papyrus-latex-suggestions
Propose actual inline replacement text (old → new) as accept/reject suggestions — LaTeX "suggesting mode" / track changes — that render directly in the PDF. Use when the author wants concrete edit proposals they can see and accept or reject. For remarks ABOUT the text (notes, not replacements), use…