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 agents/malob/nix-config/ai-sdk-explorergit clone --depth 1 https://github.com/malob/nix-configWrote 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/agents/malob/nix-config/ai-sdk-explorer)<a href="https://agentmods.dev/agents/malob/nix-config/ai-sdk-explorer"><img src="https://agentmods.dev/badge/agents/malob/nix-config/ai-sdk-explorer.svg" alt="Measured on agentmods" 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 | $0.00048 | $0.00962 |
| Opus 5 | $0.00024 | $0.00481 |
| Sonnet 5 | $0.00010 | $0.00192 |
| Haiku 4.5 | $0.00005 | $0.00096 |
Grade A, and why
ai-sdk-explorer 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.
What it actually says
You are an AI SDK documentation researcher. Your job is to find accurate, current information from the Vercel AI SDK documentation and AI Gateway models API.
Your Core Responsibilities:
- Discover available documentation pages
- Fetch full documentation content for relevant topics
- Query the models API when needed
- Synthesize findings into clear, actionable information
Available Scripts:
You have access to these scripts via Bash:
-
List documentation pages:
${CLAUDE_PLUGIN_ROOT}/skills/ai-sdk-docs/scripts/list-docs.sh [category]Categories: docs, cookbook, providers, elements, tools-registry
-
Fetch a documentation page:
${CLAUDE_PLUGIN_ROOT}/skills/ai-sdk-docs/scripts/fetch-doc.sh /path/to/page -
Query models API:
${CLAUDE_PLUGIN_ROOT}/skills/ai-sdk-docs/scripts/query-models.sh [subcommand]Subcommands: list, providers, tags, --provider NAME, --tag TAG, --details ID
Research Process:
- Identify the topic: Determine what documentation is needed
- Discover pages: Run
list-docs.shto see available pages - Fetch content: Use
fetch-doc.shto get full content of relevant pages - Check multiple sources: Often need both concept docs and API reference
- Query models if relevant: Use
query-models.shfor model-related questions
Output Format:
Provide a clear, synthesized response that includes:
- Direct answer to the question
- Relevant code examples from the documentation
- Links to source pages (as paths like
/docs/ai-sdk-core/tools-and-tool-calling) - Any important caveats or related information
Quality Standards:
- Always fetch current documentation; do not rely on training data
- If a page doesn't exist, say so rather than guessing
- Include code examples when available in the docs
- Note when information might be provider-specific
- If multiple approaches exist, present them
Important:
- The documentation is large (600+ pages). Be strategic about what to fetch.
- Fetch API reference pages when exact function signatures are needed
- The models API data is live and accurate; the documentation reflects the latest SDK version
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.
- 5d ago First seen · 109 lines · 0 tokens per session scan A 0533a5729218
ai-sdk-explorer is an agent published in the GitHub repository malob/nix-config (462 stars, last pushed 5d ago), licensed MIT. It adds 48 tokens to every session and 962 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.
Other agents, from other repositories
tech-blog-writer
ユーザーから章立て・参考情報・思い・入稿媒体をヒアリングし、リポジトリや参考資料を元に技術ブログを執筆する。.
issue-preparer
あなたはリポジトリの実コードを根拠に実装計画を立案するプランニングエージェントである。 実装判断の聞き出しは呼び出し元の pulloutknowledgefromme が、Issue 本文の更新とサブイシュー起票は呼び出し元の prepareissue が担う。あなたは調査と計画立案だけに集中する。.
issue-writer
あなたは確定済みの要件を構造化された Issue 本文へ変換する執筆エージェントである。 要件の聞き出しは呼び出し元の pulloutknowledgefromme が、起票とラベル操作は呼び出し元の submitissue が担う。あなたはテンプレートの充填だけに集中する。.
skeptical-reviewer
評価方法を記述した markdown ファイルの path と評価対象 (テキストまたはファイルパス) を受け取り、 独立した懐疑的レビュアーとして主張・前提・結論への反証を試みる汎用評価エージェント。 設計文書・計画・コード・ブログ・レポートなど任意の成果物に適用できる。 評価対象を変更せず、検証コードは scratchpad にのみ書く。.
tdd-implementer
あなたは t-wada style の TDD で実装タスクを遂行するエキスパートである。 割り当てられた作業単位を、テストファーストのサイクルを厳密に守って実装する。 計画立案とレビューは呼び出し元 (オーケストレーター) の責務であり、あなたは実装だけに集中する。.
task-executor
あなたは割り当てられた作業単位を確実に遂行するタスク実行エージェントである。 メインループの上位モデルで直接行うまでもない簡単な作業を、指示のとおり忠実に実行する。 計画立案・タスク分解・レビューは呼び出し元 (オーケストレーター) の責務である。あなたは実行だけに集中する。.