Starlight-Intelligence-System: Command for Claude Code

.claude/commands/estate-steward.md

estate-steward is a command for Claude Code from frankxai/Starlight-Intelligence-System. It costs 0 tokens per session (1,304 once invoked), scanned A, original, MIT.

An ongoing operating layer for a long-running collection of AI tools, instructions, and working data. It covers monitoring, upkeep, repairs, improvements, reporting, and support for major decisions.

In plain words
What is it for?
Use it to configure standing operations for an AI estate after its pilot or scale phase. It sets up health checks, memory cleanup, self-healing triggers, evolution suggestions, reports, and board-facilitation support.
Why use it?
It gives a built AI system regular checks after the initial setup is complete. This helps spot drift, stale information, broken connections, inefficient use, and opportunities to improve the system.

Command for Claude Code

Written for Claude Code: allowed-tools in frontmatter.

This is frankxai/Starlight-Intelligence-System's own configuration. It tells Claude Code how to work on Starlight-Intelligence-System 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 Starlight-Intelligence-System configures →

Part of the starlight-intelligence-system plugin — 6 skills, 121 commands, 7 agents shipped together

Reuse

Borrowing it

Nothing to install: this file belongs to frankxai/Starlight-Intelligence-System. 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/frankxai/Starlight-Intelligence-System/main/.claude/commands/estate-steward.md
Clone the repo
git clone --depth 1 https://github.com/frankxai/Starlight-Intelligence-System

Made for: Claude Code.

Or install starlight-intelligence-system, the plugin that ships this one along with the rest of its 6 skills, 121 commands, 7 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 estate-steward

README.md
[![agentmods](https://agentmods.dev/badge/commands/frankxai/starlight-intelligence-system/estate-steward/github.svg)](https://agentmods.dev/commands/frankxai/starlight-intelligence-system/estate-steward)
Your own site
<a href="https://agentmods.dev/commands/frankxai/starlight-intelligence-system/estate-steward"><img src="https://agentmods.dev/badge/commands/frankxai/starlight-intelligence-system/estate-steward/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 estate-steward

Your own site · 80×15
<a href="https://agentmods.dev/commands/frankxai/starlight-intelligence-system/estate-steward"><img src="https://agentmods.dev/badge/commands/frankxai/starlight-intelligence-system/estate-steward.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,304 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 $0.00000 $0.01304
Opus 5 $0.00000 $0.00652
Sonnet 5 $0.00000 $0.00261
Haiku 4.5 $0.00000 $0.00130

Measured 11d ago against content hash 1111d616ab70, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

estate-steward 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/estate-steward.md · 109 lines

How it starts

The opening of the file, as written. The whole thing — 109 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/estate-steward

Load the estate's 4-layer Blueprint, SIS-instance.md, current harness configs, and the commissioning workflow docs. Configure or invoke the Steward layer for the Standing phase (after Pilot/Scale handover).

Steward is the "run the army reliably and evolve it" role — the annuity in the factory economics. It turns a built estate into a living, compounding system the client (or their operator) can own and operate.

Input

$ARGUMENTS

Subcommands

setup (default if no subcommand)

  • Wire the production Steward primitives into the estate:
    • Health monitoring (drift detection, scheduled Proving Ground / Model Arena on the client's specific patterns, token efficiency, routing accuracy).
    • Memory hygiene (temporal half-life enforcement, contradiction sweeps, dreaming promotion, Veil audits).
    • Self-healing triggers (claws for reclamation, attestation repair, harness re-sync).
    • Evolution hooks (new module spawn suggestions, topology tweak recommendations, /sis-forge extraction points).
    • Reporting cadence (monthly health report template tied to the world-class E2E checklist).
    • Board facilitation (reminders and scaffolding for /starlight-board on major moves).
  • Create or update steward/ dir or equivalent (runbooks, dashboards config, cron/task definitions).
  • Seed the first health baseline from the Pilot phase metrics.
  • Output: Updated configs + "Steward is now active" summary with next report date.

health

  • Run a full current-state health check against the estate's Mesh + Mind.
  • Use existing claws (sentinel, attestation, memory maintenance) + orchestrator self-learning loops + evals.
  • Compare against the world-class E2E checklist (client can describe behavior, live attested work without babysitting, memory compounds, client can evolve via runbooks, gates held, measurable tech metrics).
  • Output: Health report (JSON + human summary) + any immediate remediation suggestions.

evolve

  • Propose and (with confirmation) apply evolution steps:
    • New domain sub-stack or module (via /spawn-domain-stack patterns, grounded in the estate's Genius KEEP items).
    • Topology or kernel adjustment (always surface board gate if high-stakes).
    • Harness / /si routing improvements from the upgrades track.
    • Promotion of any newly generalized patterns back to the starlight-estate-os profile.
  • Requires the estate's current 4-layer Blueprint + recent health data.
  • Output: Evolution plan + (optional) applied changes + updated Blueprint if the 4-layer shifted.

Read the full file on GitHub · 109 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 · 109 lines · 0 tokens per session scan A 1111d616ab70

Subscribe to this mod's changes

estate-steward is a command published in the GitHub repository frankxai/Starlight-Intelligence-System (8 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,304 tokens. 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.