unblocked

Answer any question about the codebase, team, or organization using Unblocked context. Use when the user asks a question that code alone can't answer — why something exists, who worked on it, what was decided, how systems connect, or what happened during an incident. Synthesizes context from PRs, docs, chat, issues, incidents, and code into a cited answer. Renders diagrams for architecture questions, tables for tradeoffs, and always cites sources.

Context-aware code review that gathers organizational context before analyzing a diff. Use when reviewing a PR or branch where team conventions, prior art, and related changes matter. Hydrates the reviewer with team patterns via context_research, categorizes findings, and presents them for discussion.

Ticket enrichment that gathers organizational context and produces findings to attach to an issue before work begins. Use when enriching a Jira, Linear, GitHub, or any issue tracker ticket with related PRs, code paths, historical context, and prior attempts. Two modes: engineering (code-focused) and support (user/product-focused). Does not solve the ticket — only enriches it.

Evidence gathering that correlates signals from code, infrastructure, and organizational sources for an incident. Use when an alert, incident, or symptom needs cross-system evidence collection — building a correlation timeline, evaluating hypotheses against evidence, and presenting findings in plain language. Presents the evidence report to the user. Does not plan the investigation or recommend fixes.

Headless code review that gathers organizational context before analyzing a diff. Use when an agent performs a PR or branch review autonomously — in CI/CD, as a sub-agent, or via `claude -p`. Hydrates the reviewer with team conventions, prior art, and related PRs via context_research, then categorizes findings as pattern mismatches, reinvented wheels, convention drift, or missing context.

Headless ticket enrichment that gathers organizational context and attaches findings to an issue before work begins. Use when an agent should autonomously enrich a Jira, Linear, GitHub, or any issue tracker ticket with related PRs, code paths, historical context, and prior attempts. Two modes: engineering (code-focused) and support (user/product-focused). Does not solve the ticket — only enriches it so whoever picks it up starts with full context.

Headless evidence gathering that correlates signals from code, infrastructure, and organizational sources for an incident. Use when an agent should autonomously gather cross-system evidence — building a correlation timeline, evaluating hypotheses against evidence, and synthesizing findings. Output is structured for AI agent consumption. Does not plan the investigation or recommend fixes.

Headless incident investigation planning that formulates multiple hypotheses from organizational context. Use when an agent should autonomously plan an investigation for an alert, incident ticket, or symptom description — correlating the affected system's architecture, recent changes, and historical incidents to produce ranked hypotheses with parallel investigation tracks. Output is structured for AI agent consumption. Does not gather evidence or recommend fixes.

Tool selection guide and legacy mapping for Unblocked search tools. Use this skill whenever you're unsure which Unblocked tool to call; when a tool call fails because a tool isn't found or isn't available; when you encounter references to deprecated tools such as data_retrieval, research_task, or unblocked_context_engine; or when a fine-grained tool like context_search_code, context_search_prs, context_query_issues, context_query_prs, or context_get_urls isn't available in your current environment. This skill explains what's available where and how to fall back gracefully.

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.

Structured, filtered issue retrieval via context_query_issues. Use this instead of context_search_issues when you need a precise filtered list — scoped by project and/or person — rather than semantic relevance matching. TRIGGER when: the user asks "what bugs are open in PROJECT", "what issues is Alice working on", "list in-progress tickets for the payments team", "issues I filed last month"; you need a definitive list rather than a ranked set of relevant items; you already know the project key or person's name and want to enumerate issues under that filter. DO NOT TRIGGER when: the question is conceptual ("is there an issue about X") or you don't have a project/person anchor — use context-search-issues for semantic matching instead.

Structured, filtered PR retrieval via context_query_prs. Use this instead of context_search_prs when you need a precise filtered list — scoped by project/repo and/or person — rather than semantic relevance matching. TRIGGER when: the user asks "what PRs did Alice open last week", "list merged PRs in payments-service this month", "PRs I'm waiting on review for", "what did the team ship yesterday"; you need a definitive list rather than a ranked set; you already know the repo name or person and want to enumerate PRs under that filter. DO NOT TRIGGER when: the question is conceptual ("why was this introduced", "what PR added feature X") — use context-search-prs for semantic matching instead.

Unified context retrieval via context_research. Searches PRs, docs, messaging platforms, issue trackers, and code history to surface WHY code works the way it does, what was tried before, and what constraints apply. TRIGGER when: you need decision history, prior art, or conventions before modifying code; planning, scoping, or researching a topic; investigating a bug or incident; you need filtered lookups across issues, messages, or PRs (e.g., "open bugs in PROJECT", "what did X work on last week"); Grep/Glob/Read failed to find the referenced code — this skill searches across repos and systems beyond the current workspace. DO NOT TRIGGER when: you only need current implementation (not history or reasoning) — use Grep/Glob/Read directly.

Code-only search across the current repo and connected repos via context_search_code. Use this instead of context_research when you want code results without Slack, Jira, or doc noise mixed in. TRIGGER when: Grep/Glob/Read can't find a referenced class, function, or module (it may live in another connected repo); you need semantic code search rather than a literal pattern match; you want to find how something is implemented or used across repos; the user asks to "find", "locate", or "search for" code. DO NOT TRIGGER when: local Grep/Glob/Read can find what you need; you also need PR history, team decisions, Slack context, or issue tracker data — use context_research instead.

Documentation-only search via context_search_documentation. Use this instead of context_research when you want docs — wikis, READMEs, runbooks, API references, architecture docs, ADRs, and onboarding guides — without Slack, PR, or issue noise mixed in. TRIGGER when: looking for setup or configuration instructions; finding a runbook or operational procedure; understanding an API contract or event schema; locating an architecture decision record or design doc; the user asks "is there docs for X", "how do I set up Y", "is this documented", or "find the runbook for Z". DO NOT TRIGGER when: you need the current code state — use Grep/Glob/Read; you need Slack threads, PRs, or issues alongside docs — use context_research instead.

Issue-only search across connected projects via context_search_issues. Use this instead of context_research when you want issue tracker results — bug reports, feature requests, tasks, and epics — without Slack, PR, or doc noise mixed in. TRIGGER when: investigating whether a bug is already known or tracked; finding the requirement or ticket behind a feature; checking if work is already planned or in progress; the user asks "is there an issue for this", "is this a known bug", "what ticket covers X", or "has this been reported before". DO NOT TRIGGER when: you need the current code state — use Grep/Glob/Read; you need Slack threads, PRs, or docs alongside issues — use context_research instead.

Query AI coding rules for this repo (extracted from CLAUDE.md, AGENTS.md, .cursorrules, .github/copilot-instructions.md, and ~40 other rules-file conventions). Invoke when the user asks about team rules, coding standards, or conventions — or when reviewing, generating, or refactoring code and you need to know which rules apply. Builds a fresh index if one doesn't exist yet.