Giovanni: Command for Claude Code

.claude/commands/calibration-report.md

calibration-report is a command for Claude Code from jaroslavsoucek-art/Giovanni. It costs 29 tokens per session (1,449 once invoked), scanned A, original, MIT.

A command that summarizes the previous month’s prediction results by actor and tier, then updates actor-scores.yaml. Here, calibration means checking how closely past predictions matched their outcomes.

In plain words
What is it for?
Use /calibration-report for the previous full month, or add --month YYYY-MM to choose a completed month.
Why use it?
It provides a regular record of prediction accuracy and prevents reports from being made from an unfinished month.

Command for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: reads .claude/ paths.

This is jaroslavsoucek-art/Giovanni's own configuration. It tells Claude Code how to work on Giovanni itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything Giovanni configures →

Reuse

Borrowing it

Nothing to install: this file belongs to jaroslavsoucek-art/Giovanni. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/jaroslavsoucek-art/Giovanni/main/.claude/commands/calibration-report.md
Clone the repo
git clone --depth 1 https://github.com/jaroslavsoucek-art/Giovanni

Made for: Claude Code.

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-report

README.md
[![agentmods](https://agentmods.dev/badge/commands/jaroslavsoucek-art/giovanni/calibration-report/github.svg)](https://agentmods.dev/commands/jaroslavsoucek-art/giovanni/calibration-report)
Your own site
<a href="https://agentmods.dev/commands/jaroslavsoucek-art/giovanni/calibration-report"><img src="https://agentmods.dev/badge/commands/jaroslavsoucek-art/giovanni/calibration-report/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.

agentmods 80×15 button for calibration-report

Your own site · 80×15
<a href="https://agentmods.dev/commands/jaroslavsoucek-art/giovanni/calibration-report"><img src="https://agentmods.dev/badge/commands/jaroslavsoucek-art/giovanni/calibration-report.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 29 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,449 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 $0.00029 $0.01449
Opus 5 $0.00015 $0.00724
Sonnet 5 $0.00006 $0.00290
Haiku 4.5 $0.00003 $0.00145

Measured 11d ago against content hash dba02a8f6ab9, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

calibration-report 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.

.claude/commands/calibration-report.md · 104 lines

How it starts

The opening of the file, as written. The whole thing — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/calibration-report

Aggregate the previous month's shadow hypothesis outcomes into a calibration report. This command is a thin shell — the aggregation logic and the > 80% accuracy guard rail live in the prediction-runtime agent. This file is the invocation contract.

Usage

/calibration-report                       # default — previous full month
/calibration-report --month <YYYY-MM>     # explicit month

Argument syntax

Arg Type Default Meaning
--month <YYYY-MM> parameterized date previous full calendar month Explicit month to aggregate. Must be in the past (no in-progress months — cohort would be incomplete).

If --month is in the future or current month → STOP: ERROR: --month <YYYY-MM> is in the future or current. Calibration aggregates completed months only.

Pre-flight (binding — STOP on failure)

Before spawning prediction-runtime:

  1. CWD check. Working directory is a Giovanni repo. Otherwise STOP.
  2. Agent definition present. .claude/agents/prediction-runtime.md exists. Missing → STOP.
  3. Calibration directories present. memory/calibration/ exists. (actor-scores.yaml is created on first run if missing.) Missing dir → STOP: ERROR: memory/calibration/ missing. The predictive layer hasn't been initialized.
  4. Month boundary enforced. If --month is omitted, the default is the previous full calendar month (today is 2026-05-21 → default is 2026-04). If today is the 1st-2nd of a month, the previous month's cohort may still have horizon spillover from the month before — surface advisory but continue.
  5. Cohort non-empty. If no shadow hypotheses match the month → STOP: INFO: no hypotheses created/resolved/expired in <YYYY-MM>. Nothing to aggregate this month. Don't write an empty report.
  6. Already-reported guard. If memory/calibration/monthly/<YYYY-MM>.md already exists, surface advisory: WARN: report for <YYYY-MM> already exists. Re-run will overwrite. Proceed? [y/N]. Default = abort.

Read the full file on GitHub · 104 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. 11d ago First seen · 104 lines · 29 tokens per session scan A dba02a8f6ab9

Subscribe to this mod's changes

calibration-report is a command published in the GitHub repository jaroslavsoucek-art/Giovanni (2 stars, last pushed 2mo ago), licensed MIT. It adds 29 tokens to every session and 1,449 once invoked, about $0.0001 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.