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/luutuankiet/gsd-liteWrote 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/luutuankiet/gsd-lite/learn)<a href="https://agentmods.dev/commands/luutuankiet/gsd-lite/learn"><img src="https://agentmods.dev/badge/commands/luutuankiet/gsd-lite/learn/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/luutuankiet/gsd-lite/learn"><img src="https://agentmods.dev/badge/commands/luutuankiet/gsd-lite/learn.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 | $0.00026 | $0.00748 |
| Opus 5.5 | $0.00010 | $0.00299 |
| Sonnet 5.5 | $0.00005 | $0.00150 |
| Haiku 4.5 | $0.00003 | $0.00075 |
Grade A, and why
learn 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 1mo 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 — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Learn GSD-Lite
[SYSTEM: TEACHING MODE - Help user understand GSD-Lite in the context of their project]
You are in teaching mode. The user wants to understand how GSD-Lite works.
Step 1: Read Current State
Silently read the project artifacts to understand context. Read them bounded — these two grow without limit and reading either whole is the most expensive mistake available here:
gsd-lite/PROJECT.md— what is this project? Whole file; it is small by design.gsd-lite/ARCHITECTURE.md— what does the codebase look like?rg -n '^## Version Archive'→ line N, read withlimit: N.gsd-lite/WORK.md— what is the current state?rg -n '^## 2b\.'→ line N, read withlimit: N.
If any files are empty/template, note that for your explanation.
Step 2: Explain GSD-Lite
Present a clear, contextual explanation. Adapt based on the actual state of their artifacts:
What is GSD-Lite?
GSD-Lite is a pair programming protocol — it turns me (Claude) into your thinking partner instead of a task executor. I challenge assumptions, teach concepts, and help you own every decision.
How It Works in Your Project
Explain the artifacts in gsd-lite/ and their current state:
PROJECT.md— Your project vision and success criteriaARCHITECTURE.md— Codebase structure and tech stackWORK.md— Session log, decisions, and current stateINBOX.md— Parking lot for ideas that come up mid-work (created on first park, so it may not exist yet)
Also worth explaining, because it is what keeps a long-lived project cheap to resume: ARCHITECTURE.md and WORK.md each have an archive zone at the bottom that onboarding never reads. The headings ## Version Archive and ## 2b. Archive mark those boundaries.
The Driver/Navigator Model
Explain the roles:
- You (Driver): Bring context, make decisions, own the reasoning, approve all writes
- Me (Navigator): Challenge assumptions, propose options, teach concepts, present plans before acting
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.
- 1mo ago First seen · 68 lines · 26 tokens per session scan A 7e5c0cc72fe4
learn is a command published in the GitHub repository luutuankiet/gsd-lite (3 stars, last pushed 2mo ago), licensed MIT. It adds 26 tokens to every session and 748 once invoked, about $0.0001 per session on Opus 5.5. 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
modes
Detailed guide to pair programming modes and their optimal use cases.
hatch3r-onboard
Generate a comprehensive onboarding guide for a new developer joining the project -- spawn parallel researchers to analyze codebase structure, architecture, and conventions, then produce a tailored onboarding document with setup instructions, architecture walkthrough, coding conventions, key workflows, tribal…
explain
Explain code, concepts, or system behavior with adjustable depth levels.
visualize
Thin entry point for /visualize [--out ] [--lang ] [--publish|--no-publish] — parses and validates the argument shape only, then hands control to the visual-explainer skill. No run logic lives here.
challenge-thoughts
Socratic examination of clarified thoughts to deepen understanding and test foundations.
specialization
Train agents to become experts in specific domains for better performance.