菜单
Retrieve clean markdown from URLs when web_fetch fails. Converts pages via Jina AI reader service with automatic retry. Use when web_fetch or curl returns 403, blocked, paywall, timeout, JavaScript-rendering errors, or empty content or user explicitly suggests using jina.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/oaustegard/claude-skills --skill fetching-blocked-urls命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Assign items to a CLOSED label vocabulary that is too large to put in a prompt — product taxonomies, category hierarchies, tag vocabularies, routing tables, ICD/SIC-style code lists. A cheap model writes the label it thinks the vocabulary would use, and an embedder snaps that writing onto the nearest legal value, so the schema is never transmitted and the output is always in-vocabulary. Use for "classify these into our taxonomy", "tag these against the existing tag list", "map these queries to categories", "the enum is too big to send", or a Literal/enum that hits a provider cap. NOT for a vocabulary that fits in a prompt — structured output measured 0.701 acc@1 there against this pattern's 0.564. NOT for open-ended labelling with no fixed vocabulary, and not for ranked retrieval over documents (bm25).
Removes LLM prose tics from drafts — staged reveals, "it's not X, it's Y", significance tags, abstraction agency, coy headers, fragment cadence, the flatter slop patterns (copula avoidance, participle tails, forced triads, chatbot residue), the flat-certainty patterns the pass itself produces (bare adverbs, juridical vocabulary, verification compounds, privative coinages), and the confiding-essayist patterns (announced honesty, stranded auxiliaries, retroactive significance, totalizing claims, obituary headlines) — and returns plain human technical prose. Use when text needs editing for register, when someone says "de-claude", "de-slop", "humanize this", "this reads like AI", "make this sound human", "remove the tics/claudisms", or asks for a voice/register pass on a post, README, report, PR description or essay. Also use before publishing any draft Claude wrote. Verifies the rewrite kept every claim. Produces either clean prose or an annotated HTML diff showing every edit with its original and reason.
Read Bluesky/ATProto without depending on Bluesky's AppView — interactions on a user's posts, thread replies, any repo's records, network-wide backlinks across every lexicon, an account's handle and PDS history, which collections are active network-wide, and layer-by-layer outage diagnosis. Use when bsky.app or the AppView is down or slow, when a Bluesky read returns timeouts or 5xx, when asked who liked/replied/quoted/reposted something, when pulling live Bluesky context cheaply, when reading records straight from a PDS, when asked who references a record or account anywhere on the network, whether an account has changed handle or migrated servers, or which lexicons and non-Bluesky apps are active on atproto. Complements browsing-bluesky, which routes everything through the AppView and fails when it does.
基于 SOC 职业分类
正在显示 SKILL.md
| name | fetching-blocked-urls |
| description | Retrieve clean markdown from URLs when web_fetch fails. Converts pages via Jina AI reader service with automatic retry. Use when web_fetch or curl returns 403, blocked, paywall, timeout, JavaScript-rendering errors, or empty content or user explicitly suggests using jina. |
| metadata | {"version":"0.1.1"} |
Retrieve readable content from URLs that web_fetch cannot access. Jina AI's reader service renders JavaScript, bypasses soft blocks, and returns clean markdown.
Invoke this skill immediately when web_fetch returns:
curl -s --max-time 30 "https://r.jina.ai/TARGET_URL"
The service returns markdown with page title, body text, and preserved links.
Jina's backend has ~10% intermittent failures. Use retry logic to achieve 99%+ success:
for attempt in 1 2 3; do
result=$(curl -s --max-time 30 "https://r.jina.ai/TARGET_URL" 2>&1)
echo "$result" | grep -q "upstream connect error" || { echo "$result"; break; }
[ $attempt -lt 3 ] && sleep 1
done
Attempt this fallback before asking users to copy-paste content manually.
Jina returns structured markdown:
Title: page titleURL Source: original URLMarkdown Content: extracted body text, links preservedr.jina.ai is whitelisted in Claude container network configuration.