calibration

calibration is a command for coding agents from aaronbassett/agent-foundry. It costs 33 tokens per session (821 once invoked), scanned A, original, MIT.

A command that asks exactly three independent estimators for a quantitative or probability-based answer, then compares how far their estimates differ.

In plain words
What is it for?
Use it for questions about timelines, delivery probabilities, costs, or other numbers where you want a confidence check.
Why use it?
It shows whether an estimate is broadly consistent or uncertain because the estimators disagree.

Command

Part of the decision-making plugin — 1 skill, 7 commands, 2 agents shipped together

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 commands/aaronbassett/agent-foundry/calibration
Clone the repo
git clone --depth 1 https://github.com/aaronbassett/agent-foundry

Or install decision-making, the plugin that ships this one along with the rest of its 1 skill, 7 commands, 2 agents.

Wrote 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.

agentmods badge for calibration

README.md
[![agentmods](https://agentmods.dev/badge/commands/aaronbassett/agent-foundry/calibration.svg)](https://agentmods.dev/commands/aaronbassett/agent-foundry/calibration)
Your own site
<a href="https://agentmods.dev/commands/aaronbassett/agent-foundry/calibration"><img src="https://agentmods.dev/badge/commands/aaronbassett/agent-foundry/calibration.svg" alt="Measured on agentmods" height="20"></a>
Per session 33 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 821 The whole file, excluding the scripts and references it only reads on demand.
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 5.1 $0.00033 $0.00821
Opus 5 $0.00016 $0.00411
Sonnet 5 $0.00007 $0.00164
Haiku 4.5 $0.00003 $0.00082

Measured 5d ago against content hash 41ff93be7471, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

calibration 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 5d 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.

plugins/decision-making/commands/calibration.md · 67 lines

How it starts

The opening of the file, as written. The whole thing — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/decision-making:calibration

When to use

The decision depends on a quantitative or probabilistic estimate, and you need a confidence check on the estimate itself — not just an answer. Examples: "how long will the migration take?", "what's the probability this ships by Q3?", "how much will the infra cost increase?"

Cost tier

Low. Exactly 3 parallel general-purpose subagents, one round. Fan-out is not configurable. See references/cost-tiers.md.

Input

A quantitative or probabilistic question + relevant context.

Why fan-out is fixed at 3

Two estimators give no triangulation signal — if they disagree, you have no way to tell which one is closer, and if they agree you can't distinguish real convergence from shared blind spots. Five or more estimators is diminishing returns for double the cost: the third estimator buys most of the signal, and each additional one buys less. Keeping the fan-out fixed at 3 forces users toward the right shape for this tool and prevents misuse as a generic parallel-thinking knob that gets dialed up whenever someone wants "more thinking."

Workflow

  1. Spawn exactly 3 parallel estimators — each general-purpose subagent is given the same question, the same context, and the same prompt. They cannot see each other's outputs. Use this exact prompt template:

    You are producing a calibrated estimate for a quantitative or
    probabilistic question.
    
    The question: [QUESTION]
    
    Context: [CONTEXT]
    
    Return three things:
    1. A point estimate (a single number with units)
    2. A confidence interval (e.g., "80% CI: 4-12 weeks")
    3. The reasoning behind your estimate (2-4 sentences explaining
       what assumptions drove it)
    
    Do not hedge with "it depends" unless you also give a specific
    number that you would commit to if forced. Do not copy the
    question back. Commit to a number.
    
    Good-faith rules: no fabricated base rates, no straw-manning the
    question, acknowledge what you're uncertain about inside your
    reasoning.
    

Read the full file on GitHub · 67 lines

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. 5d ago First seen · 67 lines · 33 tokens per session scan A 41ff93be7471

Subscribe to this mod's changes

calibration is a command published in the GitHub repository aaronbassett/agent-foundry (4 stars, last pushed 20d ago), licensed MIT. It adds 33 tokens to every session and 821 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 2026-08-31.