@specstream/pi-package 0.1.2 → 0.2.1

package/README.md

@@ -1,8 +1,13 @@
 # SpecStream for Pi
 
-Ask for any list in plain words and get every match from the live web, each row with the page it was read from: lead lists, company lists, grants, jobs, places. You see the price before anything runs and pay for each row delivered, from 5 cents each.
+The live web data engine. One platform to build, search, extract, enrich, map, and monitor data across the entire web.
 
-This is a [Pi package](https://pi.dev/packages). It adds the hosted [SpecStream](https://specstream.ai) MCP server and four skills, so Pi reads your request back, gets the price, asks you to approve it, and returns the finished list.
+Ask for any dataset in plain words, and SpecStream builds it from the entire live web: every match found, extracted and enriched into structured records, then monitored for change.
+
+- Source URL and read time on every value.
+- You see the price before anything runs and pay for each row delivered, from 5 cents each.
+
+This is a [Pi package](https://pi.dev/packages). It adds the hosted [SpecStream](https://specstream.ai) MCP server and six skills. Search the web and read pages work with no quote. For lists, Pi shows how SpecStream read your request, gets the price, asks you to approve it, and returns the rows.
 
 ## Install
 
@@ -11,18 +16,20 @@ pi install npm:@specstream/pi-package
 pi mcp login specstream
 ```
 
-The second command opens SpecStream in your browser. Sign in and approve access; Pi stores the token and refreshes it. New accounts get $10 of work free for 30 days. Inside a session, `/mcp` shows the connection and lets you sign in or out.
+The second command opens SpecStream in your browser. Sign in and approve access; Pi stores the token and refreshes it. New accounts get $10 of credit for 30 days. Inside a session, `/mcp` shows the connection and lets you sign in or out.
 
 Try it for one run without installing: `pi -e npm:@specstream/pi-package`.
 
 ## What you get
 
-- **The SpecStream MCP server** at `https://api.specstream.ai/mcp`, registered by a small extension (`extensions/specstream.ts`). It uses Pi's built-in MCP support, OAuth 2.1 with dynamic client registration, and no stored credentials. Five tools stay in view (`compile`, `quote_capacity_work`, `authorize_capacity_work`, `get_find`, `list_find_rows`); the rest load on demand through `tool_search`. In Pi they are named `mcp__specstream__<tool>`.
-- **Four skills**, loaded when the task calls for them or by name:
+- **The SpecStream MCP server** at `https://api.specstream.ai/mcp`, registered by a small extension (`extensions/specstream.ts`). It uses Pi's built-in MCP support, OAuth 2.1 with dynamic client registration, and no stored credentials. Seven tools stay in view (`search_web`, `read_pages`, `compile`, `quote_capacity_work`, `authorize_capacity_work`, `get_find`, `list_find_rows`); the rest load on demand through `tool_search`. In Pi they are named `mcp__specstream__<tool>`.
+- **Six skills**, loaded when the task calls for them or by name:
 
 | Skill | What it does |
 |---|---|
-| `/skill:specstream-compile` | Reads your request back before anything runs |
+| `/skill:specstream-search-web` | Answers a question from the live web with sources ($5 per 1,000 searches) |
+| `/skill:specstream-extract-pages` | Reads up to 20 web pages, PDFs, Word or Excel files ($0.80 per 1,000 pages) |
+| `/skill:specstream-compile` | Shows how SpecStream read your request before anything runs |
 | `/skill:specstream-review-work` | Gets the exact price and its expiry |
 | `/skill:specstream-approve-work` | Starts the list after you approve that price |
 | `/skill:specstream-continue-work` | Reads every row, sends them to Sheets, Excel, Notion, Airtable, HubSpot or Salesforce, brings back more, refreshes, monitors |
@@ -30,11 +37,7 @@ Try it for one run without installing: `pi -e npm:@specstream/pi-package`.
 ## Try it
 
 ```text
-Find every independent bookstore in Oregon with a website and a phone number.
-```
-
-```text
-List open grants for food banks in Texas, with the deadline and the award size.
+Find open grants for climate research in the EU, with the deadline and the award size.
 ```
 
 Reading a request and getting its price are free. Nothing that costs money runs until you approve an exact maximum.

package/extensions/specstream.ts

@@ -4,11 +4,11 @@ export const SERVER_NAME = "specstream";
 export const MCP_URL = "https://api.specstream.ai/mcp";
 
 const DESCRIPTION =
-  "SpecStream: ask for any list in plain words and get every match from the live web, each row with its source. Reads a request back and prices it for free; paid work needs the person's approval of an exact maximum.";
+  "SpecStream, the live web data engine. One platform to build, search, extract, enrich, map, and monitor data across the entire web. Reading and pricing a request are free; paid work needs the person's approval of an exact maximum.";
 
 // The flow that every list goes through stays visible to the model. The rest of the
 // server's tools load on demand through tool_search.
-const DIRECT_TOOLS = ["compile", "quote_capacity_work", "authorize_capacity_work", "get_find", "list_find_rows"];
+const DIRECT_TOOLS = ["search_web", "read_pages", "compile", "quote_capacity_work", "authorize_capacity_work", "get_find", "list_find_rows"];
 
 /**
  * The server entry registered with Pi. Without a key Pi signs in with OAuth (`/mcp login specstream`).

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@specstream/pi-package",
-  "version": "0.1.2",
-  "description": "SpecStream for Pi: ask for any list in plain words and get every match from the live web, each row with its source. Adds the SpecStream MCP server and four agent skills.",
+  "version": "0.2.1",
+  "description": "SpecStream for Pi. The live web data engine. One platform to build, search, extract, enrich, map, and monitor data across the entire web. Adds the SpecStream MCP server and six agent skills.",
   "type": "module",
   "keywords": [
     "pi-package",

package/skills/specstream-compile/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: specstream-compile
-description: Get the whole list from the live web by asking in plain words (leads, companies, grants, jobs, places). Reads the request back first; nothing runs or costs until a price is approved. Use when someone wants to find, build or monitor a list.
+description: Shows how SpecStream read a plain-words request, its conditions and the fields to bring back, before anything runs or costs. Use when someone wants data from the live web.
 license: MIT
 metadata:
   category: search
@@ -13,6 +13,6 @@ SpecStream is the live web data engine for teams and agents. Use the configured
 
 Use `compile` to read the person's request. Say back what it found: the kind of thing, the conditions, the details to bring back and the quantity. Separate the conditions that must hold from the details that are nice to have. Use `edit_spec` for a deliberate change. Never drop a hard condition quietly, and never treat missing evidence as a no. Reading a request does not start any work.
 
-Suggest requests that work well: local businesses with ratings, companies from registries, grants, tenders, clinical trials, papers, datasets, regulations, vulnerabilities, apps, repositories, packages, podcasts and books. Companies that use a tool, funding and new-executive signals, US nonprofits and UK financial firms by FCA status also work. If someone asks for people by job title, say that results may be few. `list_request_templates` returns requests that already work for each kind.
+If someone asks for people by job title, say that results may be few. `list_request_templates` returns requests that already work for each kind.
 
 The same saved request can be revised, branched, compared and shared, in the workspace or through this agent. Never ask for an API key in chat. Use the connector's own sign-in.

package/skills/specstream-extract-pages/SKILL.md

@@ -0,0 +1,10 @@
+---
+name: specstream-extract-pages
+description: Read the main text of up to 20 web pages, PDFs, Word or Excel files live with the Extract API ($0.80 per 1,000 pages).
+---
+
+# Read pages and documents live
+
+SpecStream is the live web data engine for teams and agents. Use the configured MCP server at `https://api.specstream.ai/mcp`.
+
+Use `read_pages` with `{"urls": ["<url>", ...]}` (1 to 20 `http` or `https` URLs) to read the clean main text, title and published date of web pages, PDFs, Word documents or Excel spreadsheets. Each page that returns text costs $0.0008 ($0.80 per 1,000 pages) from the account balance, and a page with no text is not charged. No quote or approval is needed.

package/skills/specstream-search-web/SKILL.md

@@ -0,0 +1,12 @@
+---
+name: specstream-search-web
+description: Answer a question from the live web in seconds with the Search API ($5 per 1,000 searches, no quote needed).
+---
+
+# Search the live web
+
+SpecStream is the live web data engine for teams and agents. Use the configured MCP server at `https://api.specstream.ai/mcp`.
+
+Use `search_web` with `{"q": "<the question>"}` when the person needs one fact, a fast web search with sources, or a quick look at the first rows of a request before pricing a full Deep Search run. Set `"passages": true` when you want the main text passages of the closest pages included in the reply.
+
+Each call costs $0.005 ($5 per 1,000 searches) from the account balance and needs no quote or approval. The reply returns `kind` (`fact` or `list`), `answer` (with `value`, `page` and `seen_at`), `rows` and ranked `sources`. When `kind` is `list`, `handoff` gives the estimated size and the `compile` call to price and run the full dataset.