agentpool AGENTS.md

Project instructions for AgentPool, a tool that monitors usage limits across coding-agent providers and helps send work to a provider with remaining capacity. They also define development and safety rules for the project.

In plain words
What is it for?
Use them when developing AgentPool features, provider adapters, runtime behavior, storage, tests, and documentation.
Why use it?
They keep changes consistent with the project's documented design, preserve deliberate provider choices, and prevent unsafe or unsupported behavior.

Instructions file for CodexOpenCode

Install

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.

agentmods
npx agentmods add instructions/sidduhere/agentpool/agents-md
Clone the repo
git clone --depth 1 https://github.com/sidduHERE/agentpool

Made for: Codex, OpenCode.

Per session 340 This file is loaded in full into every session.
When invoked 340 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00340 $0.00340
Opus 5 $0.00170 $0.00170
Sonnet 5 $0.00068 $0.00068
Haiku 4.5 $0.00034 $0.00034

Measured yesterday against content hash 7218dc8ee279, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agentpool AGENTS.md 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 yesterday.

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.

AGENTS.md · 29 lines

What it actually says

Talk to the user as a fellow senior developer.

AgentPool helps users get the most out of the coding-agent subscriptions they pay for: surface each provider's live usage limits and offload work to whichever still has headroom when the active provider nears its 5-hour or weekly cap. It is a control plane, not an auto-router. Keep provider/model judgment with the primary agent.

Project guardrails:

  • Treat the committed docs as the public source of truth, especially README.md, docs/architecture.md, docs/onboarding.md, docs/mcp-tools.md, and docs/usage-detection.md.
  • Use Python 3.11+.
  • Use tmux as the first runtime.
  • Preserve explicit provider selection. Do not add provider=auto.
  • Do not implement browser scraping, credential storage, model ranking, silent overage, silent merge, or silent push behavior.
  • Add fake-provider coverage before relying on real provider adapters.
  • Prefer worktree isolation for mutating tasks.
  • Use subprocess argument arrays; avoid shell strings for untrusted input.
  • Store sessions, events, usage snapshots, artifacts, and file leases in SQLite.
  • Keep artifacts outside the repo by default under ~/.agentpool/artifacts.

When AgentPool tools are available:

  1. Call inventory or usage before delegating.
  2. Choose provider/model/harness explicitly.
  3. Prefer read_only isolation for exploration and review.
  4. Use worktree isolation for edits.
  5. Spawn narrow tasks.
  6. Observe, steer, interrupt, collect, and terminate deliberately.
  7. Do not merge or push worker changes without user approval.
Changes

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.

  1. yesterday First seen · 29 lines · 340 tokens per session scan A 7218dc8ee279

Subscribe to this mod's changes

agentpool AGENTS.md is an instructions file published in the GitHub repository sidduHERE/agentpool (0 stars, last pushed 27d ago), licensed MIT. It adds 340 tokens to every session, about $0.0017 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 2026-08-31.