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-curriculum)<a href="https://agentmods.dev/commands/frankxai/starlight-intelligence-system/training-curriculum"><img src="https://agentmods.dev/badge/commands/frankxai/starlight-intelligence-system/training-curriculum/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-curriculum"><img src="https://agentmods.dev/badge/commands/frankxai/starlight-intelligence-system/training-curriculum.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 | $0.00062 | $0.02318 |
| Opus 5 | $0.00031 | $0.01159 |
| Sonnet 5 | $0.00012 | $0.00464 |
| Haiku 4.5 | $0.00006 | $0.00232 |
Grade A, and why
training-curriculum 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 5d 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 — 175 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/training-curriculum
Load agents/starlight-training.md, skills/people-intelligence/learning-architecture.md, and if present any upstream Performance Intelligence behavior-gap report or Hiring rubric for the role. Produce a Curriculum Design for the program. Hand off to exactly one next command.
Required reading at activation
- The L3 behavior change must be named, observable, and on-job. If the request is "we want training on X" without a named L3 behavior — halt. Ask once. If still not named, refuse and route upstream to
/performance-reviewor the Org/Culture sub-system. - The L4 business outcome must be named. "Engagement will go up" is not L4; "1:1 quality score on Q3 engagement survey lifts ≥0.5 vs Q2 baseline" is L4.
Input
$ARGUMENTS
Flags
--behavior "<observable on-job behavior change>"— required. Kirkpatrick L3.--outcome "<business metric tied to behavior>"— required. Kirkpatrick L4.--audience <role | cohort>— required. e.g.people-managers,IC-engineers-L4-L5,customer-success-team-EMEA.--duration <N weeks>— required. The curriculum spans this window inclusive of spacing checkpoints.--cohort-size <N>— optional. Affects modality mix (cohort-based vs self-paced).--prereq <slug>— optional. Links to a prior program whose curriculum is prerequisite knowledge.
Process
-
Verify outcome inputs. If
--behavioror--outcomeis missing or vague, halt and ask once. Do not generate curriculum from "general topic" framing. Outcome-first or refuse. -
Reverse-engineer capability. From the L3 behavior, list the sub-skills required. From sub-skills, list the prerequisite knowledge. From prerequisite knowledge, identify what the audience already has vs. what must be loaded. The capability map is the spine of the curriculum.
-
Sequence the spine per the protocol in
learning-architecture.md:- Concept introduction (light)
- Application practice (with feedback)
- Spaced retrieval at R1 (1d), R2 (1w), R3 (1m), R4 (3m)
- Simulation in performance-context
- On-job application (with manager reinforcement)
- Measurement (L3 at 30/60/90; L4 at 90+)
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.
- 5d ago First seen · 175 lines · 62 tokens per session scan A 226a4269104a
training-curriculum is a command published in the GitHub repository frankxai/Starlight-Intelligence-System (8 stars, last pushed today), licensed MIT. It adds 62 tokens to every session and 2,318 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.
agent-list
List all available agents with their capabilities (MPM-provided — may be shadowed by a user /agent-list command).