From unblocked-skills
Direct URL content resolution via context_get_urls. Use this to fetch full content from one or more URLs the agent already has in hand — PRs, issues, docs, Slack messages, or arbitrary web pages that were surfaced by a prior search or provided by the user. TRIGGER when: you already have concrete URLs from a previous context_research / context_search_* result and want the full body, not just the title and URL; the user pasted one or more links and wants them summarized or analyzed; you need to resolve a Jira/Linear/PR/Slack link directly. DO NOT TRIGGER when: you need to search for content (use context_research or the context_search_* family); the URL is local (localhost, file://, private network) — those aren't reachable.
How this skill is triggered — by the user, by Claude, or both
Slash command
/unblocked-skills:unblocked-context-get-urlsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Direct URL resolution. Calls `context_get_urls` with one or more URLs to return their full content — bypassing semantic search when the agent already knows exactly which documents it wants.
Direct URL resolution. Calls context_get_urls with one or more URLs to return their full content — bypassing semantic search when the agent already knows exactly which documents it wants.
Sources: any URL reachable through Unblocked's configured connectors (PRs, issues, docs, messages) and arbitrary public web pages.
context_get_urls is exposed on both CLI and MCP — unlike the context_search_* and context_query_* family, you will see it in your MCP tool list. Prefer the CLI when available for uniform behavior. Run command -v unblocked once per session and cache the result. See unblocked-tools-guide for full routing rules.
CLI (preferred):
unblocked context-get-urls --urls "<url1>" "<url2>" ...
MCP fallback (use if CLI is confirmed unavailable): call context_get_urls with urls as an array. Exposed on MCP in virtually all environments.
If neither is available: stop and tell the user Unblocked is not configured in this environment (see unblocked-tools-guide for the full message). Do not substitute with a generic web-fetch tool for private connector resources (Jira, Linear, Slack, private GitHub) — those require Unblocked's auth.
If the URL is truly public (a blog post, public docs) and you don't need auth, a plain fetch also works — but this tool handles both cases.
| Situation | Use |
|---|---|
| User pasted 2 Jira links and asked "summarize these" | context-get-urls |
Prior context_research returned 5 PR URLs and you want their full bodies | context-get-urls |
| You want to find PRs related to rate limiting | context-search-prs (or context-research) |
| You want to understand why a change was made, with supporting citations | context-research, then context-get-urls on the top result |
| You need to re-open a specific known link and act on its content | context-get-urls |
context-get-urls is a follow-up tool. Search first, then resolve the most relevant URLs if the titles + snippets aren't enough.
| Parameter | Required | Description |
|---|---|---|
urls | Yes | One or more URLs to retrieve content from. CLI accepts multiple space-separated values: --urls <url1> <url2>. MCP expects an array. |
Good inputs:
https://acme.atlassian.net/browse/PAY-1234https://linear.app/acme/issue/ENG-567https://github.com/acme/payments-service/pull/89https://github.com/acme/payments-service/issues/42https://acme.slack.com/archives/C0123/p1712345678Bad inputs (will fail or return nothing useful):
localhost, 127.0.0.1, private network addressesfile:// URLshttp:///https://)The exact set of supported URL patterns is org-specific — it depends on which connectors (GitHub orgs, Jira/Linear workspaces, Slack, Notion, etc.) have been configured for the current user. Before assuming a URL will resolve, inspect the live list of patterns from the tool itself:
unblocked context-get-urls --help. The output includes a "Supported URL patterns" section listing every regex that the current installation accepts (e.g. which GitHub orgs are wired up, which Slack workspace, which Jira/Linear/Notion hosts).context_get_urls tool description exposed by the MCP server. The same per-org pattern list is embedded in the tool's description field, so tools/list (or the equivalent in the agent's MCP client) will surface it.If a user-supplied URL doesn't match any listed pattern, either the connector isn't configured or the URL belongs to a different org/tenant — stop and tell the user instead of calling the tool and getting an empty result.
When you have a related cluster of URLs, pass them all in one call rather than making one call per URL. Example: the top 3 PRs returned from a context-research pass.
unblocked context-get-urls \
--urls "https://github.com/acme/payments-service/pull/89" \
"https://github.com/acme/payments-service/pull/91" \
"https://acme.atlassian.net/browse/PAY-1234"
Keep batches reasonably sized (roughly 5–10 URLs); very large batches can time out or exceed output limits.
gh pr diff locally if you need the changeNo separate references directory — usage is narrow enough that this SKILL.md is self-contained.
Guides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.
npx claudepluginhub unblocked/skills --plugin unblocked-skills