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 zorost/AI-Engineering-Lab --skill agent-ops-handoffgit clone --depth 1 https://github.com/zorost/AI-Engineering-LabWrote 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/zorost/ai-engineering-lab/agent-ops-handoff)<a href="https://agentmods.dev/skills/zorost/ai-engineering-lab/agent-ops-handoff"><img src="https://agentmods.dev/badge/skills/zorost/ai-engineering-lab/agent-ops-handoff/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/zorost/ai-engineering-lab/agent-ops-handoff"><img src="https://agentmods.dev/badge/skills/zorost/ai-engineering-lab/agent-ops-handoff.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00044 | $0.00992 |
| Opus 5 | $0.00022 | $0.00496 |
| Sonnet 5 | $0.00009 | $0.00198 |
| Haiku 4.5 | $0.00004 | $0.00099 |
Grade A, and why
agent-ops-handoff 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 9d 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 — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Ops Handoff
1 · Purpose
Close the gap between "the agent works on my machine" and "the agent runs unattended and someone owns it", tracing, cost control, scheduling, alerting, and a way back.
2 · When to use
- Moving any agent from interactive use to scheduled, event-driven, or user-facing operation (including personal assistants like OpenClaw on a VPS).
- When an unattended agent has surprised its owner, the post-incident checklist.
3 · Inputs
- The agent with loop safety in place (
agent-loop-safety). - The run schedule or trigger (cron, webhook, message gateway).
- An owner: a named human who gets the alerts.
4 · Procedure
- Centralize traces. Write per-run traces (thoughts, tool calls, costs) to a durable store, not scrollback. Queryable by run ID, date, and status.
- Stand up the cost dashboard. Per-run dollars and tokens, daily rollups, and the budget line. Cost per successful task is the metric; raw spend is the alarm.
- Schedule deliberately. Cron or the platform scheduler with three rules: off-peak minutes (not :00/:30), an explicit timezone, and a catch-up policy for missed runs written down (skip, do not stampede).
- Alert on the four signals. Run failure, cost-cap trip, guardrail refusal, and silence (a scheduled run that never reported). Route to the owner, not a channel nobody reads.
- Gate the outputs. Unattended writes to shared systems (email, tickets, production tables) stay human-approved until 30 clean runs. Count them.
- Write the rollback. How to disable the schedule, revoke the credentials, and restore the prior version, as commands, tested once.
- Define the weekly review. 30 minutes: read 5 traces (2 failures, 3 random), review cost per task, check guardrail counts. This is the external feedback loop made operational.
- Hand over. The owner acknowledges the dashboard, the alerts, and the rollback, in writing (an issue comment is fine).
5 · Anti-rationalization
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.
- 9d ago First seen · 91 lines · 44 tokens per session scan A d34d8d53d99d
agent-ops-handoff is a skill published in the GitHub repository zorost/AI-Engineering-Lab (309 stars, last pushed 22d ago), licensed MIT. It adds 44 tokens to every session and 992 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 skills, from other repositories
fine-tuning-os
Use this skill when conducting a professional LLM fine-tuning engagement in Zero-Data mode via the MCP server fine-tuning-os. Triggers on: prestation de fine-tuning LLM, livraison de modèle affiné, pipeline Zero-Data, orchestration des 64 outils fine-tuning-os, cycle de vie entraînement (préparation / données…
ai-engineering-toolkit
6 production-ready AI engineering workflows: prompt evaluation (8-dimension scoring), context budget planning, RAG pipeline design, agent security audit (65-point checklist), eval harness building, and product sense coaching.
llm-ops
LLM Operations -- RAG, embeddings, vector databases, fine-tuning, prompt engineering avancado, custos de LLM, evals de qualidade e arquiteturas de IA para producao.
llama-factory
Expert guidance for fine-tuning LLMs with LLaMA-Factory - WebUI no-code, 100+ models, 2/3/4/5/6/8-bit QLoRA, multimodal support.
tinker-training-cost
Calculates training costs for Tinker fine-tuning jobs. Use when estimating costs for Tinker LLM training, counting tokens in datasets, or comparing Tinker model training prices. Tokenizes datasets using the correct model tokenizer and provides accurate cost estimates.
transformers
This skill should be used when working with pre-trained transformer models for natural language processing, computer vision, audio, or multimodal tasks. Use for text generation, classification, question answering, translation, summarization, image classification, object detection, speech recognition, and fine-tuning…