Giovanni: Command for Claude Code

.claude/commands/shadow-review.md

shadow-review is a command for Claude Code from jaroslavsoucek-art/Giovanni. It costs 25 tokens per session (2,038 once invoked), scanned A, original, MIT.

A quarterly review command for checking past shadow hypotheses, which are tracked predictions kept for later evaluation. It reviews unresolved predictions past their target date and samples recently resolved ones.

In plain words
What is it for?
It is for auditing prediction records by date, actor, time window, or review sample size.
Why use it?
It creates a repeatable lookback process for checking whether predictions were handled according to governance rules.

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/shadow-review.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 shadow-review

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

Your own site · 80×15
<a href="https://agentmods.dev/commands/jaroslavsoucek-art/giovanni/shadow-review"><img src="https://agentmods.dev/badge/commands/jaroslavsoucek-art/giovanni/shadow-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 25 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,038 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 $0.00025 $0.02038
Opus 5 $0.00013 $0.01019
Sonnet 5 $0.00005 $0.00408
Haiku 4.5 $0.00003 $0.00204

Measured 12d ago against content hash 038f90132ad2, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

shadow-review 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 12d 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/shadow-review.md · 122 lines

How it starts

The opening of the file, as written. The whole thing — 122 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/shadow-review

Process unresolved past-horizon shadow hypotheses and audit a sample of recently resolved ones for governance compliance. This command is a thin shell — the adversarial lookback discipline (binding principle 7) lives in the prediction-runtime agent. This file is the invocation contract.

Usage

/shadow-review                          # default — all past-horizon pending + 10-20 resolved sample (90d window)
/shadow-review --sample 20              # custom sample size
/shadow-review --actor <slug>           # focus on hypotheses touching one actor
/shadow-review --window <YYYY-MM>       # focus on one month's resolved cohort
/shadow-review --horizon <YYYY-MM-DD>   # review hypotheses with horizon_at <= date
/shadow-review --force-review-now       # override COI guard — recent resolutions excluded from cohort

Argument syntax

Arg Type Default Meaning
--sample <N> parameterized int 10-20 (auto) Size of the resolved sample to audit. Caller can override for deeper review.
--actor <slug> parameterized string none Filter the cohort to hypotheses whose actor field matches <slug>. Useful when calibration-report flagged an actor with concerning patterns.
--window <YYYY-MM> parameterized last 90 days Focus on a single calendar month's resolved cohort instead of rolling 90-day window.
--horizon <YYYY-MM-DD> parameterized date today Review hypotheses whose horizon_at <= <date>. Default = today. Useful for pre-audit dry runs ("what would shadow-review have closed if I'd run it last week?").
--force-review-now boolean flag off Override the COI guard. Hypotheses with resolved_date < 48 h ago are excluded from the cohort rather than reviewed — the override unblocks the run, it does not admit the conflicted files.

All three filter args (--actor, --window, --horizon) can combine.

Pre-flight (binding — STOP on failure)

Before spawning prediction-runtime:

Read the full file on GitHub · 122 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. 12d ago First seen · 122 lines · 25 tokens per session scan A 038f90132ad2

Subscribe to this mod's changes

shadow-review is a command published in the GitHub repository jaroslavsoucek-art/Giovanni (2 stars, last pushed today), licensed MIT. It adds 25 tokens to every session and 2,038 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.