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 valtterimelkko/agent-workflow-skills --skill long-horizon-waiting-strategiesgit clone --depth 1 https://github.com/valtterimelkko/agent-workflow-skillsWrote 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/valtterimelkko/agent-workflow-skills/long-horizon-waiting-strategies)<a href="https://agentmods.dev/skills/valtterimelkko/agent-workflow-skills/long-horizon-waiting-strategies"><img src="https://agentmods.dev/badge/skills/valtterimelkko/agent-workflow-skills/long-horizon-waiting-strategies/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/valtterimelkko/agent-workflow-skills/long-horizon-waiting-strategies"><img src="https://agentmods.dev/badge/skills/valtterimelkko/agent-workflow-skills/long-horizon-waiting-strategies.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 | $0.00206 | $0.09136 |
| Opus 5.5 | $0.00082 | $0.03654 |
| Sonnet 5.5 | $0.00041 | $0.01827 |
| Haiku 4.5 | $0.00021 | $0.00914 |
Grade A, and why
long-horizon-waiting-strategies scanned grade A 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 21d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
your behalf by a repo script, curl command or `pi-orch watch register`; that serves How it starts
The opening of the file, as written. The whole thing — 628 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Long-horizon waiting strategies
How to wait for hours without burning context, holding a turn open, or — the failure that actually bites — never waking up again.
What to read, in three tiers
Tier 1 — Must read, every run that will idle. These named sections, in
order, before you end any turn on a wait: Harness applicability (which wake
path applies to you), The pre-idle checklist, The child contract (every
dispatch mandates orchestrated-child-worker), Choose the backstop for the failure you
are covering and The backstop ladder.
Tier 2 — Must read when the situation matches. Conditional references — mandatory within their situation, skippable outside it:
| Reference | Mandatory when… |
|---|---|
references/supervision-and-stuck-detectors.md |
you will dispatch or supervise a child, or build any stuck-detector — read before the dispatch, not after |
references/claude-code-watch-wake-and-background-watcher.md |
you are a Claude Code CLI parent — before that harness's first long wait |
references/antigravity-reactive-watcher-and-schedule.md |
you are an Antigravity CLI/IDE parent — before that harness's first long wait (covers wait-watch.sh and orchestrator invariants) |
Tier 3 — Situational. Supervising a dispatched child (near the end: operator-confirmed capability, over-supervision, the liveness discriminator, the hand-back protocol, goal cases) is required reading before you dispatch an agent child and skippable when you are waiting on a build, a measurement window, or a batch job with no child agent.
Skipping a Tier 1 or Tier 2 item is how idle windows end with nobody waking up.
The problem this solves
Any wait longer than a few minutes has three bad options and one good one:
- Polling in-turn (check, sleep, check again inside your own turn) burns context, blocks the session, and dies with the turn.
- Holding the turn open on a stream or a
waitcall dies with the connection. - Ending your turn with no wake mechanism means the work finishes and nobody notices — you sit idle until a human asks "did it work?". This is the worst outcome and it happens silently.
- The good option: end your turn with a verified wake path and backstop, go fully idle, and reconcile when the wake arrives.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 21d ago Changed · +10 lines ef5ca5cc937a
- 22d ago Changed · +12 lines 11a835d14f3e
- 24d ago First seen · 606 lines · 206 tokens per session scan A 6f46f9f15e80
long-horizon-waiting-strategies is a skill published in the GitHub repository valtterimelkko/agent-workflow-skills (2 stars, last pushed 21d ago), licensed MIT. It adds 206 tokens to every session and 9,136 once invoked, about $0.0008 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-01.
Other skills, from other repositories
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
component-fixtures
Use when creating or updating component fixtures for screenshot testing or their shared infrastructure, or when designing UI components to be fixture-friendly. Covers fixture file structure, theming, service setup, CSS scoping, async rendering, validation, and common pitfalls.