{
  "markdown": "# agentcents — LLM cost consulting skill for Claude\n\nA [Claude Skill](https://agentskills.io) that turns Claude into a neutral LLM\ncost advisor: estimate what any call or agent workflow will cost across\ncandidate models, get the **cheapest model that clears the task's capability\nbar**, and see the comparison as a Mermaid diagram — with honest ranges, never\nfalse point estimates.\n\nWorks two ways:\n\n- **Consult mode (no install):** answers from a bundled, hand-verified\n  snapshot of current model prices and Intelligence Index scores.\n- **Live mode:** if the [agentcents](https://pypi.org/project/agentcents/)\n  proxy is running locally, the skill uses your real ledger — calibrated\n  forecasts, spend, and accuracy instead of priors.\n\n## Install\n\n**Claude.ai / Desktop:** Settings → Capabilities → Skills → Add → upload this\nfolder (or the packaged `.skill`).\n\n**Claude Code:** `/plugin marketplace add labham-llc/agentcents-skill`\nthen `/plugin install agentcents@labham-skills`\n\nThen just ask things like:\n\n> what would it cost to classify ~500 support emails a day, about 200 tokens each?\n>\n> I'm building a 5-step research agent — which model should each step use?\n\n## What's in the box\n\n```\nSKILL.md                 — the consulting method (measure, range, bar, pick, flag instability)\nassets/pricing.json      — price + capability snapshot (see _meta.verified_utc)\nreferences/local-api.md  — live-mode reference for a locally running agentcents\n```\n\n**Snapshot verified: 2026-07-15** (see `_meta.verified_utc`) — refreshed with each agentcents release.\n\n## The product behind it\n\n[agentcents](https://labham.com/agentcents.html) is a local proxy that meters,\nforecasts, and budget-enforces LLM API calls — and grades its own forecasting\naccuracy on a calibration page. `pip install agentcents`. The skill is free;\nso is the agentcents free tier.\n\n© Labham LLC · skill instructions and snapshot data may be redistributed with\nattribution; the agentcents software itself is proprietary (see its LICENSE).\n",
  "bytes": 2012,
  "sha": "817852aaf0fe497e698c2fb7bebf38256058f40c2cf6f6ebe00f8d3ce4a4170e",
  "repo_slug": "labham-llc/agentcents-skill",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_labham_llc_agentcents_14302496/readme"
}