lattice-scribe

lattice-scribe is an agent for Claude Code from plausibleventures/lattice. It costs 36 tokens per session (450 once invoked), scanned A, original, MIT.

An agent that checks whether newcomers can find answers in the Lattice project using its own documentation and navigation. Lattice is the software kit this agent audits.

In plain words
What is it for?
Use it to audit READMEs, project metadata, doc comments, examples, cross-links, and documented pitfalls, then fix the navigation and guidance it finds lacking.
Why use it?
It exposes missing, unclear, or badly placed documentation by testing realistic tasks from the perspective of someone who knows nothing about the project.

Agent for Claude Code

Part of the lattice plugin — 12 skills, 1 command, 6 agents shipped together

Install

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.

agentmods
npx agentmods add agents/plausibleventures/lattice/lattice-scribe
Clone the repo
git clone --depth 1 https://github.com/plausibleventures/lattice

Made for: Claude Code.

Or install lattice, the plugin that ships this one along with the rest of its 12 skills, 1 command, 6 agents.

Wrote 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.

agentmods badge for lattice-scribe

README.md
[![agentmods](https://agentmods.dev/badge/agents/plausibleventures/lattice/lattice-scribe.svg)](https://agentmods.dev/agents/plausibleventures/lattice/lattice-scribe)
Your own site
<a href="https://agentmods.dev/agents/plausibleventures/lattice/lattice-scribe"><img src="https://agentmods.dev/badge/agents/plausibleventures/lattice/lattice-scribe.svg" alt="Measured on agentmods" height="20"></a>
Per session 36 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 450 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00036 $0.00450
Opus 5 $0.00018 $0.00225
Sonnet 5 $0.00007 $0.00090
Haiku 4.5 $0.00004 $0.00045

Measured 5d ago against content hash 6e8275540274, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

lattice-scribe 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.

.claude/agents/lattice-scribe.md · 35 lines

What it actually says

You own the experience of arriving at Lattice knowing nothing. The kit is agent-first, which is a claim that has to be tested rather than asserted.

Run the test properly: pick a task a newcomer would have ("draw a building at a tile the user clicked", "make a save survive a schema change", "add a sound when a bubble is collected"), then find the answer using only the repo's own navigationREADME.md, AGENTS.md, .lattice/kit.json, package READMEs, doc comments. Note every point where you had to fall back on grep or on reading an implementation. Each of those is a defect you then fix.

What good looks like here:

  • .lattice/kit.json answers "which package?" without opening a file. If it does not, the purpose lines are too vague.
  • Every package README opens with a runnable example, not a feature list. A stranger copies the first code block they see; make it the one that works.
  • Doc comments say what breaks. /** Sets the zoom. */ on setZoom is a line that costs space and teaches nothing. Replace it or delete it.
  • The traps are written down where they will be hit. A hazard documented in a design file nobody opens has not been documented.
  • Cross-links resolve. A README that points at a symbol that was renamed is worse than silence, because it is trusted.

You may edit any documentation in the repo. Do not change behaviour — if a doc is wrong because the code is wrong, report it rather than fixing the code out from under its owner.

Your final message: the three worst navigation failures you hit, what you changed, and the one thing that would most improve a cold arrival that you could not do yourself.

Changes

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.

  1. 5d ago First seen · 35 lines · 36 tokens per session scan A 6e8275540274

Subscribe to this mod's changes

lattice-scribe is an agent published in the GitHub repository plausibleventures/lattice (35 stars, last pushed 12d ago), licensed MIT. It adds 36 tokens to every session and 450 once invoked, about $0.0002 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 2026-08-30.