Borrowing it
Nothing to install: this file belongs to lucyfuur94/webnav-core. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/lucyfuur94/webnav-core/main/CLAUDE.mdgit clone --depth 1 https://github.com/lucyfuur94/webnav-coreWrote 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/instructions/lucyfuur94/webnav-core/claude-md)<a href="https://agentmods.dev/instructions/lucyfuur94/webnav-core/claude-md"><img src="https://agentmods.dev/badge/instructions/lucyfuur94/webnav-core/claude-md/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/instructions/lucyfuur94/webnav-core/claude-md"><img src="https://agentmods.dev/badge/instructions/lucyfuur94/webnav-core/claude-md.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 | $0.09398 | $0.09398 |
| Opus 5.5 | $0.03759 | $0.03759 |
| Sonnet 5.5 | $0.01880 | $0.01880 |
| Haiku 4.5 | $0.00940 | $0.00940 |
Grade A, and why
webnav-core CLAUDE.md 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 2mo 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 — 236 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — webnav-core / webnav
Read this first, every session. These are settled decisions. Do not deviate without an explicit decision to change them.
Fix upstream, never patch downstream (settled 2026-07-10 — HARD RULE)
When something wrong shows up in a late stage (the graph, the dashboard, a walk), that is a SYMPTOM, not the bug. Find the earliest stage in the pipeline that let it in and fix it there — recording → effects → draftFromEffects → build → graph → viewer. A fix that scrubs/filters/special-cases the bad thing after it's already in the data is FORBIDDEN — it leaves every other consumer of that data still wrong and hides the real defect. Correct the producing logic so the symptom disappears everywhere at once, then re-verify from the true source (re-run the pipeline end-to-end, not just re-render) and close. If you can't reach the upstream source, say so explicitly rather than patching downstream. (User, repeated: "we can't patch downstream, fix logic upstream which solves downstream automatically and reverify and close.")
Structure inference (settled 2026-07-10 — 2026-07-10-structure-inference-design.md)
Recording→graph derives structure by OBSERVATION-BASED INFERENCE, never one-shot site-tuned heuristics. Five site-agnostic evidence axes: (1) settledness — only readiness-gated landings define a state's face; redirect-chain URLs are aliases, never states; (2) declaration — ARIA landmarks declare shell, dialog/menu/listbox declare overlays; a click INSIDE an overlay is a value selection, not a page affordance; (3) cross-page variance — nodes on ≥80% of pages = shell, stored once on the _shell state (edges project from-anywhere); (4) cross-visit variance — a state's core = what repeats across its landings; single-landing states are provisional and analyse REPORTS what to record next; (5) within-page repetition — ≥3 same-role siblings sharing a name suffix fold to ONE scope:'row' affordance. Identity = URL-template × structural-template (propose/dispose). The map stores structure, never data values (no option lists, no instance headings, no counts/dates). Enforced by tests/guidelines.test.ts.
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.
- 2mo ago First seen · 236 lines · 9,398 tokens per session scan A 49623bf63d63
webnav-core CLAUDE.md is an instructions file published in the GitHub repository lucyfuur94/webnav-core (0 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 9,398 tokens to every session, about $0.0376 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.
webnav-core CLAUDE.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| webnav-core CLAUDE.mdlucyfuur94/webnav-core | Instructions file | 0 | 1mo ago | Apache-2.0 | A | 9,398 | Claude Code |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 196k | today | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 144k | today | MIT | A | 8,130 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 196k | today | MIT | A | 5,685 | GitHub Copilot |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 150k | today | MIT | A | 4,680 | Codex, OpenCode |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 283k | today | MIT | A | 4,342 | Codex, OpenCode |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 196k | today | MIT | A | 3,153 | GitHub Copilot |