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 code-yeongyu/senpi --skill bungit clone --depth 1 https://github.com/code-yeongyu/senpiWrote 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/code-yeongyu/senpi/bun)<a href="https://agentmods.dev/skills/code-yeongyu/senpi/bun"><img src="https://agentmods.dev/badge/skills/code-yeongyu/senpi/bun/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/code-yeongyu/senpi/bun"><img src="https://agentmods.dev/badge/skills/code-yeongyu/senpi/bun.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 | $0.00237 | $0.02758 |
| Opus 5.5 | $0.00095 | $0.01103 |
| Sonnet 5.5 | $0.00047 | $0.00552 |
| Haiku 4.5 | $0.00024 | $0.00276 |
Grade A, and why
bun scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
| HTTP/2 + HTTP/3 server; fetch protocol/compress/proxy; per-client pools | `http2: true`, `http3: true`, `fetch(url, { protocol, compress, proxy })`, `new Bun.FetchSession()` | undici Agent | http-networking | How it starts
The opening of the file, as written. The whole thing — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Bun 1.4.x — Use the Builtins First
Bun 1.4.0 (Aug 2026) through 1.4.4 (Oct 2026) ships builtins and CLIs that replace most utility npm packages and several dev tools. Before adding a dependency, spawning a tool, or writing a workaround, check the map below — if Bun ships it, use the builtin. Version tags ([v1.4.3]) in the references give exact minimums; dated notes (2026-10-25) mark facts measured in our own sessions. Release posts: https://bun.com/blog/bun-v1.4 … https://bun.com/blog/bun-v1.4.4.
Capability map — what you can do now
| You want to... | Use | Replaces | Ref = references/<name>.md |
|---|---|---|---|
Type-check a TS project: tsc-identical errors, 3-6x faster, no typescript install |
bun check --no-script, bun run/test/build --check |
tsc, tsgo | dev-tooling |
| Format with your Prettier/oxfmt config | bun fmt --no-script (--check, --changed, --staged) |
prettier, oxfmt, lint-staged | dev-tooling |
Run shell commands from JS, no /bin/sh |
Bun.$`...` → .nothrow().quiet().text()/.json()/.lines() |
execa, zx, shelljs | session-recipes |
| Spawn with timeout / PTY / IPC; measure a child | Bun.spawn({ env: process.env, timeout, killSignal, terminal, ipc }), proc.resourceUsage() |
node-pty, tree-kill, pidusage | session-recipes |
| Keep tokens out of files, argv and env dumps | Bun.secrets.get/set (OS keychain), --env-file=/dev/stdin |
keytar, dotenv | session-recipes |
| Resize/convert/rotate images | Bun.file(p).image().resize().webp().write() |
sharp | runtime-apis |
| Headless browser: navigate, click, screenshot, evaluate, CDP | new Bun.WebView() |
puppeteer, playwright (scraping) | runtime-apis |
| Scrape or rewrite HTML by CSS selector, streaming | new HTMLRewriter().on("a[href]", { element }) |
cheerio, jsdom (scraping) | runtime-apis |
| Markdown → HTML / React / ANSI | Bun.markdown.html() / .react() / .ansi() |
marked, react-markdown, glow | runtime-apis |
| Schedule cron jobs | Bun.cron() |
node-cron | runtime-apis |
| Run package scripts concurrently, glob-matched | bun run --parallel "build:*" |
concurrently, npm-run-all | runtime-apis |
| Parse JSON5 / JSONL / JSONC / XML / TOML / YAML; tarballs; globs | Bun.JSON5/.JSONL/.JSONC/.XML/.TOML/.YAML, Bun.Archive, Bun.Glob |
json5, ndjson, jsonc-parser, fast-xml-parser, @iarna/toml, js-yaml, tar, fast-glob | runtime-apis |
| ANSI-aware terminal text; table rendering | Bun.stringWidth/sliceAnsi/wrapAnsi/stripANSI(), Bun.inspect.table() |
string-width, slice-ansi, wrap-ansi, strip-ansi, cli-table | runtime-apis |
| Hash, UUID v7, semver, deep-equal, gzip/zstd, which | Bun.hash/CryptoHasher/password, randomUUIDv7, Bun.semver, deepEquals, gzipSync/zstdCompressSync, Bun.which |
uuid, semver, lodash.isequal, pako, which | runtime-apis |
| Postgres / MySQL / SQLite, Redis, S3-compatible (R2) clients | Bun.SQL, bun:sqlite, Bun.RedisClient, Bun.S3Client / s3:// |
pg, mysql2, better-sqlite3, ioredis, aws-sdk | runtime-apis |
| Serve a static dir (ETag/Range/304 handled) | Bun.serve({ routes: { "/x/*": { dir } } }) |
express.static, serve-static, sirv | http-networking |
| HTTP/2 + HTTP/3 server; fetch protocol/compress/proxy; per-client pools | http2: true, http3: true, fetch(url, { protocol, compress, proxy }), new Bun.FetchSession() |
undici Agent | http-networking |
| Stream a download to disk; replay a request as curl | Bun.write(path, await fetch(url)), BUN_CONFIG_VERBOSE_FETCH=curl |
pipeline + createWriteStream | http-networking |
| Tests across CPU workers / CI shards / changed files / flaky retry | bun test --parallel --shard=1/3 --changed=main --retry |
jest/vitest infra | test |
| Kill "passes alone, fails in suite" bugs; hunt flakes | bun test --isolate, --rerun-each N, --only-failures |
vitest default behavior | test |
| Single-file executables with embedded assets and profile-guided startup; feature flags; in-memory bundling; metafile; React Compiler | bun build --compile --asset ./public --bytecode-order, bun:bundle feature(), files:{}, metafile: true, --react-compiler |
pkg hacks, esbuild define/plugins, babel-plugin-react-compiler | build |
| Diff package versions, fix vulns, dedupe/prune, licenses, explain a dep | bun pm diff, bun audit fix, bun dedupe, bun prune, bun pm licenses, bun why |
npm-diff, npm audit fix, npm ls | install |
| Offline / cached installs; Docker layers; CI cache; 7x faster CI | bun install --offline|--prefer-offline, bun pm fetch, bun pm cache pack|unpack, linker = "isolated" |
pnpm store | install |
| Profile CPU/heap/bundle as grep-able Markdown | bun --cpu-prof-md, --heap-prof-md, bun build --metafile-md |
Chrome DevTools round-trip | dev-tooling |
Run untrusted JS without eval / new Function / FFI |
--disallow-code-generation-from-strings=strict, --no-ffi-cc |
vm2 | security |
Run Playwright, vitest, Next.js 16, OpenTelemetry, dd-trace; undici, node-fetch, ws |
they now just work (the last three load Bun's builtin even when installed) | Node.js | node-compat-platforms |
What ships with it
11 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.
- references/breaking-changes.md 8.4 KB
- references/build.md 8.1 KB
- references/dev-tooling.md 5.1 KB
- references/http-networking.md 8.9 KB
- references/install.md 11 KB
- references/node-compat-platforms.md 4.2 KB
- references/performance.md 3.4 KB
- references/runtime-apis.md 12 KB
- references/security.md 7.1 KB
- references/session-recipes.md 8.6 KB
- references/test.md 9.6 KB
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 · 54 lines · 237 tokens per session scan A 5504854ef8a6
bun is a skill published in the GitHub repository code-yeongyu/senpi (505 stars, last pushed today), licensed MIT. It adds 237 tokens to every session and 2,758 once invoked, about $0.0009 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-11-01.
Other skills, from other repositories
Manage emails via Himalaya CLI. Use when the user wants to read, send, search, or organize emails. Triggers on "email", "inbox", "send email", "check mail", "reply to", "forward", or mentions of Gmail, Outlook, iCloud, Proton Mail.
install-jazz-github-action
Set up Jazz as a GitHub Action for automated PR code review and on-demand /jazz PR assistance. Use when the user asks to add Jazz to CI, create a GitHub Action workflow, set up PR review automation, enable /jazz commands on PRs, or integrate Jazz with GitHub.
create-workflow
Create Jazz workflow automation files (WORKFLOW.md). Use this for scheduling Jazz agents to run recurring tasks. For OS-level scripts/commands, use create-system-routine.
obsidian
Use the official Obsidian CLI (v1.12+) to manage vaults, notes, daily notes, search, tasks, tags, properties, links, templates, sync, publish, and workspaces. Use when the user mentions Obsidian, vaults, notes, or wants to automate note-taking and knowledge base workflows.
skill-creator
Create new Jazz skills for automating workflows. Use when the user asks to create a skill, make a skill, or wants to define custom automation behavior.
meeting-notes
Summarize meetings and extract action items. Use when processing meeting transcripts, notes, or recordings to get summaries, decisions, and follow-ups.