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/training-scenarios)<a href="https://agentmods.dev/commands/frankxai/starlight-intelligence-system/training-scenarios"><img src="https://agentmods.dev/badge/commands/frankxai/starlight-intelligence-system/training-scenarios/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/training-scenarios"><img src="https://agentmods.dev/badge/commands/frankxai/starlight-intelligence-system/training-scenarios.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 | $0.00075 | $0.02085 |
| Opus 5 | $0.00037 | $0.01043 |
| Sonnet 5 | $0.00015 | $0.00417 |
| Haiku 4.5 | $0.00007 | $0.00209 |
Grade A, and why
training-scenarios 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 — 165 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/training-scenarios
Load agents/starlight-training.md, skills/people-intelligence/learning-architecture.md, and (if it exists) the Hiring sub-system's structured-interview rubric pattern for the same skill domain. Produce a Scenario Library — case-based, decision-rich, rubric-scored.
Why scenarios
Slide-based training fails encoding-specificity (Tulving) by construction: the context at encoding (a slide deck in a hotel ballroom) does not match the context at retrieval (an actual customer call, an actual feedback conversation, an actual code review). Scenarios bridge that gap.
A good scenario:
- Sets up an on-job situation with enough context to feel real and enough ambiguity to require judgment.
- Forces decisions — the learner must choose, not just observe.
- Has a rubric for response quality — not "right" or "wrong" but "what does a strong response include vs. a weak one."
- Includes a debrief guide — the discussion that converts the scenario from a quiz into a learning experience.
Format mirrors the Hiring sub-system's structured-interview rubric — same discipline applied to skill-development.
Input
$ARGUMENTS
Flags
--difficulty-levels <comma-separated>— required. Typical:intro,mid,advanced. Drives stratification across the library.--count <10-15>— required. Library size. Below 10, retrieval-practice variety is insufficient. Above 15, the library bloats and the discipline of "every scenario must matter" weakens.--context <on-job context>— required. The actual situation the scenarios mirror. e.g., "1:1 feedback conversation with a struggling direct report," "customer-success renewal-risk call," "code review of junior engineer's PR."--prior-program <slug>— optional. If scenarios are for a specific program, link the scenarios to module sequence.
Process
- Verify domain + context specificity. A scenario library on "communication" is useless. A library on "1:1 feedback conversations with struggling direct reports in a remote-first context" is useful. If the domain is too vague, halt and re-scope.
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 · 165 lines · 75 tokens per session scan A dd1488460717
training-scenarios is a command published in the GitHub repository frankxai/Starlight-Intelligence-System (8 stars, last pushed today), licensed MIT. It adds 75 tokens to every session and 2,085 once invoked, about $0.0004 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.
agent-list
List all available agents with their capabilities (MPM-provided — may be shadowed by a user /agent-list command).