Search
Run one natural-language query that fans out across connected chat, email, storage, tracker, CRM, and knowledge-base MCP tools and returns cited, synthesized answers.
Overview
Search is a journey-wide agent skill that queries every connected MCP source in one pass—usable whenever a solo builder needs to locate a decision, document, or discussion before committing to more work.
Install
npx skills add https://github.com/anthropics/knowledge-work-plugins --skill searchWhat is this skill?
- Decomposes vague questions into 2–4 targeted sub-queries before searching
- Runs parallel searches across chat, email, cloud storage, project tracker, CRM, and knowledge-base MCP tools
- Synthesizes ranked results with source, date, and excerpt in a consistent table-style presentation
- Surfaces next steps when results are thin or a single authoritative source dominates
- Redirects to CONNECTORS.md when no MCP sources are connected
- Targets 2–4 decomposed sub-queries per user question
- Documents six common MCP source families (chat, email, cloud storage, project tracker, CRM, knowledge base)
Adoption & trust: 2.8k installs on skills.sh; 19.6k GitHub stars; 3/3 security scanners passed (skills.sh audits).
What problem does it solve?
You know a decision, spec, or thread exists somewhere across chat, email, or docs, but hunting each connector manually wastes agent context and breaks flow.
Who is it for?
Solo builders with multiple MCP connectors who need fast, cited recall across tools during implementation, reviews, or ops firefighting.
Skip if: Builders with zero MCP sources connected, or searches that need deep code-index semantics inside a single repo without external connectors.
When should I use this skill?
User says “find that doc about…”, “what did we decide on…”, “where was the conversation about…”, or needs a decision, document, or discussion that may live in any connected MCP source.
What do I get? / Deliverables
You get a concise synthesis with per-source excerpts, dates, and follow-up suggestions so you can act on the finding or narrow the query without re-searching each tool by hand.
- Short synthesized answer with ranked findings
- Per-source results with title, date, excerpt, and link or reference
- Suggested follow-up queries or single-source deep-dive when coverage is thin
Recommended Skills
Journey fit
Useful at every journey phase - explore requirements and options before committing to a direction.
Where it fits
Pull earlier competitor or market notes that live in email archives and a shared drive folder before committing to a niche.
Recover the exact scope or pricing thread from chat and a project tracker before locking a prototype milestone.
Locate the API credential doc and the Slack decision on vendor choice while wiring MCP-backed integrations.
Find launch checklist comments and stakeholder sign-off messages scattered across chat and email before release day.
Trace when an incident workaround was agreed in chat versus what was filed in the knowledge base before changing production behavior.
How it compares
Use instead of opening each chat, inbox, and drive separately—the skill is parallel MCP orchestration, not a single-service grep or a static skills.sh search.
Common Questions / FAQ
Who is search for?
Search is for solo and indie builders who connect knowledge-work MCP tools and want one command to hunt decisions, documents, and threads across those sources with cited results.
When should I use search?
Use search when you need scattered recall—during idea research for prior notes, validate for past scope decisions, build for specs in storage or chat, ship for launch threads, or operate when debugging against historical decisions.
Is search safe to install?
It only queries MCP connectors you already authorized; review the Security Audits panel on this Prism page and limit connectors to data you are comfortable exposing to your agent.
SKILL.md
READMESKILL.md - Search
# Search Command > If you see unfamiliar placeholders or need to check which tools are connected, see [CONNECTORS.md](../../CONNECTORS.md). Search across all connected MCP sources in a single query. Decompose the user's question, run parallel searches, and synthesize results. ## Instructions ### 1. Check Available Sources Before searching, determine which MCP sources are available. Attempt to identify connected tools from the available tool list. Common sources: - **~~chat** — chat platform tools - **~~email** — email tools - **~~cloud storage** — cloud storage tools - **~~project tracker** — project tracking tools - **~~CRM** — CRM tools - **~~knowledge base** — knowledge base tools If no MCP sources are connected: ``` To search across your tools, you'll need to connect at least one source. Check your MCP settings to add ~~chat, ~~email, ~~cloud storage, or other tools. Supported sources: ~~chat, ~~email, ~~cloud storage, ~~project tracker, ~~CRM, ~~knowledge base, and any other MCP-connected service. ``` ### 2. Parse the User's Query Analyze the search query to understand: - **Intent**: What is the user looking for? (a decision, a document, a person, a status update, a conversation) - **Entities**: People, projects, teams, tools mentioned - **Time constraints**: Recency signals ("this week", "last month", specific dates) - **Source hints**: References to specific tools ("in ~~chat", "that email", "the doc") - **Filters**: Extract explicit filters from the query: - `from:` — Filter by sender/author - `in:` — Filter by channel, folder, or location - `after:` — Only results after this date - `before:` — Only results before this date - `type:` — Filter by content type (message, email, doc, thread, file) ### 3. Decompose into Sub-Queries For each available source, create a targeted sub-query using that source's native search syntax: **~~chat:** - Use available search and read tools for your chat platform - Translate filters: `from:` maps to sender, `in:` maps to channel/room, dates map to time range filters - Use natural language queries for semantic search when appropriate - Use keyword queries for exact matches **~~email:** - Use available email search tools - Translate filters: `from:` maps to sender, dates map to time range filters - Map `type:` to attachment filters or subject-line searches as appropriate **~~cloud storage:** - Use available file search tools - Translate to file query syntax: name contains, full text contains, modified date, file type - Consider both file names and content **~~project tracker:** - Use available task search or typeahead tools - Map to task text search, assignee filters, date filters, project filters **~~CRM:** - Use available CRM query tools - Search across Account, Contact, Opportunity, and other relevant objects **~~knowledge base:** - Use semantic search for conceptual questions - Use keyword search for exact matches ### 4. Execute Searches in Parallel Run all sub-queries simultaneously across available sources. Do not wait for one source before searching another. For each source: - Execute the translated query - Capture results with metadata (timestamps, authors, links, source type) - Note any sources that fail or return errors — do not let one failure block others ### 5. Rank and Deduplicate Results **Deduplication:** - Identify the same information appearing across sources (e.g., a decision discussed in ~~chat AND confirmed via email) - Group related results together rather than showing duplicates - Prefer the most authoritative or complete version **Ranking factors:** - **Relevance**: How well does the result match t