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 skills add spuentesp/lain --skill lain-architecturegit clone --depth 1 https://github.com/spuentesp/lainWrote 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/skills/spuentesp/lain/lain-architecture)<a href="https://agentmods.dev/skills/spuentesp/lain/lain-architecture"><img src="https://agentmods.dev/badge/skills/spuentesp/lain/lain-architecture/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/skills/spuentesp/lain/lain-architecture"><img src="https://agentmods.dev/badge/skills/spuentesp/lain/lain-architecture.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 | $0.00060 | $0.00571 |
| Opus 5.5 | $0.00024 | $0.00228 |
| Sonnet 5.5 | $0.00012 | $0.00114 |
| Haiku 4.5 | $0.00006 | $0.00057 |
Grade A, and why
lain-architecture 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 19d 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 — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Lain architecture — map the system
The arch package answers shape questions the core deliberately
does not. Load it when the question is about structure rather than
one symbol.
Load it
Call load_package("arch") (or set LAIN_TOOL_PROFILE=arch at
config time). If the tools do not appear in your list, reconnect
the client and refetch. The raw package (load_package("raw"))
adds the low-level layer.
The tools
explore_architecture— file/module tree to a chosen depth. Start here for "what does this subsystem look like".compare_modules— stability and coupling of two modules side by side. Use for "which of these do we refactor first".architectural_observations— patterns, boundary violations, high-fan-out hot spots. Use for architecture review, not for one change.trace_dependency— everything a symbol depends on, recursively. The upstream question; the downstream one isget_blast_radius.get_layered_map/get_context_depth— slices of the stack from an entry point; how deep a symbol sits.navigate_to_anchor/get_anchor_score— climb from a leaf to its controlling anchor; how load-bearing one symbol is.get_master_map— staleness: when each module last synced. Read this before trusting a stale-looking answer.suggest_refactor_targets— complexity + stability hot spots.
raw package, when the high-level tools are not enough:
query_graph (JSON ops-array against the graph — read
describe_schema first), get_code_snippet, get_call_sites
(every call with its line), explain_symbol,
get_context_for_prompt, get_cross_runtime_callers,
semantic_search (needs a model).
Rules of thumb
- Depth fields take a number (
3) or a range ("1..3"). - A layered map is for shape; do not use it to answer "who calls
this" — that is
get_call_chain. query_graphis the escape hatch, not the first tool. Its schema is indescribe_schema; prefer the named tools unless you need a traversal nobody anticipated.
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.
- 19d ago First seen · 53 lines · 60 tokens per session scan A 920799c0ea49
lain-architecture is a skill published in the GitHub repository spuentesp/lain (9 stars, last pushed 2d ago), licensed MIT. It adds 60 tokens to every session and 571 once invoked, about $0.0002 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 2026-09-28.
Other skills, from other repositories
lsp-mcp-server
Use with Codex, Claude Code, or another MCP client when lsp-mcp-server is connected and semantic code navigation, analysis, refactoring, or diagnostics are needed. Covers all 29 lsp tools for TypeScript, Python, Rust, Go, C/C++, Ruby, PHP, Elixir, Kotlin, Java, and user-configured languages. Prefer these tools to text…
explore
Build an accurate mental model of an unfamiliar codebase, feature, or area before changing it — where it lives, how it connects, what it does, and why. Use when asked how something works, where something is, when onboarding to a repo, or before editing code you don't know. For a cross-cutting question that needs…
debug
Trace a bug, error, or regression to its root cause. Use when investigating a failure, a stack trace, unexpected behavior, or "why does X happen".
speclite-review-adversarial-general
A general review skill that critically examines code, specifications, stories, diffs, or documents for gaps, contradictions, risks, and unclear claims. It returns a Markdown list of findings.
speclite-review-edge-case-hunter
A review skill that traces code branches and lists only edge cases the code does not handle. It checks inputs such as diffs, files, or functions and returns findings as JSON.
debug-ci
Use when CI/CD pipeline fails. Provides systematic diagnosis for lint, type, test, and build failures across Python, TypeScript, Java, Rust, and n8n SDKs.