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.
git clone --depth 1 https://github.com/frankxai/Starlight-Intelligence-SystemWrote 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.
[](https://agentmods.dev/commands/frankxai/starlight-intelligence-system/sound-composition-demo)<a href="https://agentmods.dev/commands/frankxai/starlight-intelligence-system/sound-composition-demo"><img src="https://agentmods.dev/badge/commands/frankxai/starlight-intelligence-system/sound-composition-demo/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.
<a href="https://agentmods.dev/commands/frankxai/starlight-intelligence-system/sound-composition-demo"><img src="https://agentmods.dev/badge/commands/frankxai/starlight-intelligence-system/sound-composition-demo.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00050 | $0.01144 |
| Opus 5 | $0.00025 | $0.00572 |
| Sonnet 5 | $0.00010 | $0.00229 |
| Haiku 4.5 | $0.00005 | $0.00114 |
Grade A, and why
sound-composition-demo 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 7d 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.
How it starts
The opening of the file, as written. The whole thing — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/sound-composition-demo
Load verticals/sound-intelligence/SKILL.md, agents/starlight-sound-composition.md, skills/sound-intelligence/composition-architecture.md, and the song's Score / Lyric / Arrangement architectures. Produce a Demo Plan.
Disclaimer (non-waivable)
Demos that incorporate samples, AI tools, or co-writer contributions touch rights territory. This is system architecture, not legal advice.
Input
$ARGUMENTS
Process
- Disclaim.
- Read upstream. Score / Lyric / Arrangement architecture.
- What does this demo prove? Different demos prove different things — vocal melody and lyric / harmonic motion and form / arrangement viability / tempo and feel. Naming what's being proven matters more than scope.
- What does this demo NOT need to prove yet? Equally important. Demos that try to prove everything become production sessions and lose the proof.
- Capture method matched to proof. One-take vocal-and-instrument / rough-multitrack-on-DAW / near-final-arrangement-with-placeholder-mix / boombox-on-piano. Method serves the proof, not the other way.
- Reference-track grounding. 2-3 reference tracks the demo is in conversation with. Not to copy — to position.
- Time budget. Demos sized to fit the cycle (a 2-hour demo for a vocal-melody proof; a 1-day demo for arrangement-viability proof; never a 2-week demo unless the demo is the song).
- Save.
sound-intelligence/composition/demo-<song-slug>-<YYYY-MM-DD>.md. - Hand off. Default: capture the demo (not a command — actual studio time). Then:
/sound-production-mix-planif proof passes.
Output format
# Demo Plan — <Song Title> — <YYYY-MM-DD>
> **Demos involving samples, AI tools, or co-writer contributions touch rights. This is system architecture, not legal advice.**
## Context
- Song / Score / Lyric / Arrangement references: <paths>
## What this demo proves
<One specific thing. Not "the song" — the song is too big for a demo. The vocal-and-melody. Or the harmonic-and-form. Or the arrangement-viability. Or the tempo-and-feel.>
## What this demo does NOT need to prove yet
<Equally important. List what is explicitly out of scope for this demo so the demo can be a demo and not a production session.>
## Capture method
- **Method:** <one-take vocal-and-instrument | rough-multitrack | near-final-arrangement-with-placeholder-mix | boombox-on-piano>
- **Why this method serves the proof:** <one paragraph>
- **Time budget:** <e.g., 2 hours / 1 day / 1 weekend>
## Reference tracks
1. <Track 1 — what it positions us against>
2. <Track 2 — what it positions us against>
3. <Track 3 — what it positions us against (optional)>
Reference tracks position; they do not template. The demo is in conversation with them, not derivative of them.
## Pass / fail criteria
<What does this demo passing look like? What does it failing look like? Specific.>
## Load-bearing next move
**Capture the demo within the time budget.** Then: if proof passes → `/sound-production-mix-plan <song-slug>`; if proof fails → revisit upstream architecture (Score / Lyric / Arrangement).
---
**Built on SIP** — Starlight Intelligence Protocol
- Substrate: starlightintelligence.org/protocol v1.1.0
- Layers used: [file-contract, attestation, commands, sovereignty]
- Verticals: [email protected] (Sound Intelligence reference vertical — Composition sub-system)
- Generated: <ISO date>
---
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.
- 7d ago First seen · 99 lines · 50 tokens per session scan A c19d5ba3468e
sound-composition-demo is a command published in the GitHub repository frankxai/Starlight-Intelligence-System (8 stars, last pushed today), licensed MIT. It adds 50 tokens to every session and 1,144 once invoked, about $0.0003 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 2026-09-03.
Other commands, from other repositories
memories
View and manage learned memories.
mpm-session-resume
Load context from paused session.
forget
Delete specific memories.
learn
Add new learning to memory.
gbu-retro
Post-session retrospective — harvest this session's lessons into durable doctrine.
agentic-jujutsu
The learning backbone of ACOS. Based on ruvnet's agentic-jujutsu (v2.3.6, MIT), customized for ACOS.