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 khalilbenaz/claude-skills-collection --skill pool-managergit clone --depth 1 https://github.com/khalilbenaz/claude-skills-collectionWrote 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/khalilbenaz/claude-skills-collection/pool-manager)<a href="https://agentmods.dev/skills/khalilbenaz/claude-skills-collection/pool-manager"><img src="https://agentmods.dev/badge/skills/khalilbenaz/claude-skills-collection/pool-manager/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/khalilbenaz/claude-skills-collection/pool-manager"><img src="https://agentmods.dev/badge/skills/khalilbenaz/claude-skills-collection/pool-manager.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 | $0.00094 | $0.02635 |
| Opus 5 | $0.00047 | $0.01318 |
| Sonnet 5 | $0.00019 | $0.00527 |
| Haiku 4.5 | $0.00009 | $0.00264 |
Grade A, and why
pool-manager 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 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.
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 — 280 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Pool Manager
Quand utiliser un pool d'agents
| Situation | Pool recommandé ? |
|---|---|
| Latence de cold-start > 500 ms (chargement modèle, init tools) | Oui |
| Rafales de tâches homogènes (même type d'agent) | Oui |
| Tâches ponctuelles, types variés et imprévisibles | Non |
| Budget CPU/mémoire contraint, peu de concurrence | Non — instancier à la demande |
| SLA strict sur le temps de réponse (< 200 ms P95) | Oui |
Règle rapide : si le coût de warm-up dépasse 20 % du temps de traitement moyen d'une tâche, un pool est rentable.
Workflow en 10 étapes
1. Définir la topologie du pool
Choisir entre pool statique (taille fixe, simple, prévisible) et pool dynamique (auto-scaling, plus complexe).
- Pool statique : cas d'usage à charge constante, environnements embarqués.
- Pool dynamique : SaaS, charges variables, pics prévisibles ou non.
Identifier les types d'agents et leur proportion :
researcher_pool : min=3, max=10
coder_pool : min=2, max=8
reviewer_pool : min=1, max=4
2. Initialiser le pool au démarrage
Pré-créer min_size agents, effectuer un health check avant de les marquer available.
async def initialize(self):
for _ in range(self.min_size):
agent = await self._create_agent()
if await agent.ping(): # health check initial
self._all_agents[agent.id] = agent
await self._available.put(agent)
else:
await agent.destroy() # ne pas injecter un agent cassé
print(f"[POOL:{self.agent_type}] {self._available.qsize()} agents prêts")
3. Checkout avec timeout obligatoire
Ne jamais attendre indéfiniment. Lever une exception explicite plutôt que bloquer.
async def _checkout(self, timeout: float = 5.0) -> PooledAgent:
try:
agent = await asyncio.wait_for(self._available.get(), timeout=timeout)
except asyncio.TimeoutError:
raise PoolExhaustedError(
f"Aucun agent {self.agent_type} disponible après {timeout}s "
f"(pool size={len(self._all_agents)}, waiters={self._waiters})"
)
agent.status = "busy"
agent.use_count += 1
return agent
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 · 280 lines · 94 tokens per session scan A 64ebfeecfe28
pool-manager is a skill published in the GitHub repository khalilbenaz/claude-skills-collection (22 stars, last pushed 17d ago), licensed MIT. It adds 94 tokens to every session and 2,635 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
stripe-payments
Stripe integration covering Checkout, Payment Intents, subscriptions, webhook verification, idempotency, and SCA.
stripe-link-cli
Agent payments via Stripe Link — cards, SPT, approvals.
asc-ppp-pricing
Set territory-specific pricing for subscriptions and in-app purchases using current asc setup, pricing summary, price import, and price schedule commands. Use when adjusting prices by country or implementing localized PPP strategies.
ai-slop
Operational rubric that turns "don't make AI slop" into observable properties, severity levels, evidence requirements, and repair actions for interface design. Use as the reference rubric when building or reviewing marketing sites, product interfaces, dashboards, portfolios, or e-commerce pages, especially alongside…
ecommerce-patterns
E-commerce: cart, checkout, payments (Stripe/Adyen), order state, inventory, promos, tax. Triggers: cart, checkout, SKU, payment, Stripe, Shopify, Medusa, Magento, coupon, refund.
amazon-returns-recovery
Suede-affiliated Amazon money-recovery audit for restocking fees, short or denied refunds, and Amazon-billed subscriptions. Use when the user mentions a restocking fee, a short or denied Amazon refund, or a forgotten Prime Video Channel, Audible, Kindle Unlimited, or Prime charge, or asks whether Amazon still owes or…