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/reclaim-knowledge)<a href="https://agentmods.dev/commands/frankxai/starlight-intelligence-system/reclaim-knowledge"><img src="https://agentmods.dev/badge/commands/frankxai/starlight-intelligence-system/reclaim-knowledge.svg" alt="Measured on agentmods" 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 | $0.00058 | $0.03346 |
| Opus 5 | $0.00029 | $0.01673 |
| Sonnet 5 | $0.00012 | $0.00669 |
| Haiku 4.5 | $0.00006 | $0.00335 |
Grade A, and why
reclaim-knowledge 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 8d 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 — 215 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/reclaim-knowledge
Load SIP.md, VOICES.md, and genius/profile-<person-slug>.md if it exists (its frameworks list will inform the domain structure). Also load skills/intelligence/knowledge-reclamation.md and agents/starlight-genius.md if they exist in this release. If either is missing, emit a one-line notice: knowledge-reclamation skill / genius agent not yet loaded — proceeding with architect voice + Frank DNA and continue.
Take a person's scattered professional material — Canva, Google Drive, local folders, Notion, Dropbox, screenshot archives, old company laptops — and organize it into a coherent second-brain architecture. Structure is by function (what you DO), not by source (where it happened to live). Output is a single Reclamation Map the person can execute in a 3-hour drag-and-drop session, no terminal required.
Input
$ARGUMENTS
When this command fires
- The person has material scattered across 3+ platforms and cannot answer "where does X live" without opening half of them.
- They want a second brain they can actually find things in — not a new tool to add, a sorting rule for what they already have.
- Typically runs alongside or right after
/discover-genius— the Genius Profile's frameworks are the natural top-level taxonomy. - Can also run standalone if the person knows their functional domains already (rare but valid).
When this command does NOT fire
- When the ask is really about picking a new tool. This command does not recommend Notion vs Obsidian vs Drive. It organizes within whatever the person already uses.
- When nothing is scattered — one tool, clean folders, easy to find. Route to
/discover-geniusinstead; they already have the shape, they need the excavation. - When the person wants someone else to do it for them. The person must do the drag-and-drop. This command produces the map; execution is theirs.
Process
-
Source inventory. Ask the person — verbatim, because asking specifically matters:
"Name every place your work currently lives. Every one. Include the ones you're avoiding. Canva account. Google Drive folders (personal AND any company accounts you still have access to). Local Desktop and Documents folders. Notion pages. Dropbox. Old company laptops you can still log into. Email attachments. Screenshots folder on your phone and computer. Anything you'd grieve losing.
For each source, rough size: number of files or estimated hours of work inside. No need to count precisely — eyeballed orders of magnitude are fine."
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.
- 8d ago First seen · 215 lines · 58 tokens per session scan A 56b3a45eb8c4
reclaim-knowledge is a command published in the GitHub repository frankxai/Starlight-Intelligence-System (8 stars, last pushed today), licensed MIT. It adds 58 tokens to every session and 3,346 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 2026-08-31.
Other commands, from other repositories
agent
Create and manage custom AI agents.
memories
View and manage learned memories.
mpm-session-resume
Load context from paused session.
learn
End-of-session learning loop — analyze signals, approve instincts, capture directives, refresh MEMORY.md. Invokes dreamteam learn; one source of behavioral truth shared with the team.md SESSION LEARNING step.
forget
Delete specific memories.
learn
Add new learning to memory.