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.
npx agentmods add commands/frankxai/starlight-intelligence-system/map-relationshipsgit 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/map-relationships)<a href="https://agentmods.dev/commands/frankxai/starlight-intelligence-system/map-relationships"><img src="https://agentmods.dev/badge/commands/frankxai/starlight-intelligence-system/map-relationships.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 | $0.00059 | $0.02298 |
| Opus 5 | $0.00030 | $0.01149 |
| Sonnet 5 | $0.00012 | $0.00460 |
| Haiku 4.5 | $0.00006 | $0.00230 |
Grade A, and why
map-relationships 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 6d 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 — 177 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/map-relationships
Load SIP.md, ALLIANCE.md, VOICES.md, agents/starlight-relational.md, skills/relational/network-architecture.md. Map the person's network into a living architecture with audit, gaps, leverage points, alliance candidates (surfaced only — not forged), investment plan, and cadence. Personal. Never public.
Input
$ARGUMENTS
Process
-
Ingest. Identify the person. Collect the initial roster of 20–40 names from
$ARGUMENTS. If fewer than 20 names are provided, halt and return:"Need at least 20 named people in your work year to architect a network. Anything less is an abstraction. Start here: who did you interact with on work this week? This month? This quarter? Build the list up from there. Come back when you have 20+ names."
Do not proceed without a real inventory.
-
Sort by type. Assign every name exactly one relationship type: peer, mentor, mentee, client, collaborator, family, acquaintance. Overlaps collapse to the dominant type for this person's work right now. Name each person's one-line context (how they know the person) and most recent interaction.
-
Strength audit. For every name, capture: frequency of contact (weekly / monthly / quarterly / yearly / dormant), depth of trust (high / medium / low), value flow direction (compounds / drains / balanced / one-way gift). If zero drains surface, re-audit — real networks have at least one.
-
Leverage identification. Name the top 3–5 compounding leverage points. Introduction graph, referral flows, collaboration compounding, expertise borrowing. Specific, not generic.
-
Gap-scan. Name missing roles as concrete archetypes (not abstractions). Examples: "no mentor in [named domain]", "no peer at current stage", "no mentee", "no protocol-defender voice in the network", "no collaborator with skill X." Each gap becomes a specific search, not a wish.
-
Alliance-surface (pre-filter only). From compounding relationships only, flag candidates where all four forging conditions (
ALLIANCE.md) could hold. Surface names; do not forge. For each surfaced candidate, recommend/design-alliance-readiness <candidate-name>as the next step. Most architectures surface 0–1 candidates. Zero is valid.
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.
- 6d ago First seen · 177 lines · 59 tokens per session scan A 524098f5c9b8
map-relationships is a command published in the GitHub repository frankxai/Starlight-Intelligence-System (8 stars, last pushed yesterday), licensed MIT. It adds 59 tokens to every session and 2,298 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
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.
consolidate
Write a compact checkpoint summary of the current frontier.