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 agentmods add skills/friedbotstudio/baseline/code-browsernpx skills add friedbotstudio/baseline --skill code-browsergit clone --depth 1 https://github.com/friedbotstudio/baselineWhat 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 | $0.00201 | $0.04249 |
| Opus 5 | $0.00101 | $0.02124 |
| Sonnet 5 | $0.00040 | $0.00850 |
| Haiku 4.5 | $0.00020 | $0.00425 |
Grade A, and why
code-browser 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 yesterday.
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 — 194 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Code Browser
A natural code-traversal skill. Answers "where does this come from?" by walking from the entry point downward, not by string-searching for the name.
It is the primary path for code-navigation questions in any language (CLAUDE.md Article X.5): the universal walk is the first attempt; reach for the Explore agent or grep only when the repo has no resolvable structure or the walk dead-ends. Pure full-text search and direct type/util definition lookups stay with grep — they are not navigation.
When this skill exists
The user-visible label of a thing (a third-party integration's icon, a "Refresh" button, a panel header) is rarely the right keyword to grep for. The same word often appears in unrelated flows — e.g., grepping a vendor's product name can land on an unrelated documents API while the icon itself is rendered by a generic integrations registry that fetches a different endpoint. The reliable path is always top-down: start at the page, follow imports, stop at the network boundary. This skill encodes that walk and supplies a fast helper so it costs essentially nothing.
The universal walk
This works on any codebase, regardless of framework:
- Locate the entry point for the user-visible thing. For web: the page/route file. For a single-page app, the screen component. The user's question almost always names the page directly or implicitly (e.g. "the loan-application page").
- Read the entry point and follow imports downward. Note components rendered (JSX, route children), data hooks called (
useFoo()), and service/API helpers invoked. - Stop at a network/IO boundary: a
fetch/axios/SDK call, a SQL query, a file read. The URL or query you find there is the answer. - Verify: open the leaf file once with
Readand confirm the URL/symbol is still what the cache claims.
If at any point you find yourself globally grepping for the user-visible label, stop — you've left the walk.
Optional accelerator (JS/TS only): walk.mjs
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- yesterday First seen · 194 lines · 201 tokens per session scan A 393fd14cf355
code-browser is a skill published in the GitHub repository friedbotstudio/baseline (11 stars, last pushed 5d ago), licensed Apache-2.0. It adds 201 tokens to every session and 4,249 once invoked, about $0.0010 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.
Other skills, from other repositories
dev-standards
Enforces development workflows, quality gates, coding standards, and release processes for the deterministic-agent-control-protocol project. Use when implementing features, fixing bugs, refactoring architecture, adding integrations, updating policies, writing tests, updating documentation, or preparing releases.
code-review-with-lsp
Code review with LSP-powered code intelligence. Uses MCP tools (diagnostics, hover, references, definition, symbols) for semantic code understanding, not just text grep.
i18n-check
国际化完整性检查。检查翻译 key 是否缺失、硬编码文本、locale 文件一致性。.
vue-best-practices
Vue 2/3 代码规范检查。包括组件命名、Props 校验、Composition API 规范等。.
python-review
Python 遗留代码审查：bare except、SQL 注入、反序列化、密钥、调试输出.
rust-review
Rust 服务审查：panic、SQL 注入、密钥、错误吞没、遗留标记.