@apmantza/greedysearch-pi 1.0.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -0,0 +1,12 @@
1
+ {
2
+ "permissions": {
3
+ "allow": [
4
+ "Bash(gh api:*)",
5
+ "Bash(python -c \"import sys,json,base64; open\\(''''/tmp/pi-tools-ex.ts'''',''''w'''',encoding=''''utf-8''''\\).write\\(base64.b64decode\\(json.load\\(sys.stdin\\)[''''content'''']\\).decode\\(''''utf-8''''\\)\\)\")",
6
+ "Read(//tmp/**)",
7
+ "Bash(for f:*)",
8
+ "Bash(mkdir:*)",
9
+ "Bash(node:*)"
10
+ ]
11
+ }
12
+ }
package/README.md ADDED
@@ -0,0 +1,59 @@
1
+ # GreedySearch for Pi
2
+
3
+ Pi extension that adds a `greedy_search` tool — fans out queries to Perplexity, Bing Copilot, and Google AI simultaneously and returns synthesized AI answers.
4
+
5
+ Forked from [GreedySearch-claude](https://github.com/apmantza/GreedySearch-claude).
6
+
7
+ ## Install
8
+
9
+ ```bash
10
+ pi install npm:@apmantza/greedysearch-pi
11
+ ```
12
+
13
+ Or directly from git:
14
+
15
+ ```bash
16
+ pi install git:github.com/apmantza/GreedySearch-pi
17
+ ```
18
+
19
+ ## Usage
20
+
21
+ Once installed, Pi gains a `greedy_search` tool. The model will use it automatically for questions about current libraries, error messages, version-specific docs, etc.
22
+
23
+ You can also invoke it directly:
24
+
25
+ ```
26
+ greedy_search({ query: "best way to handle auth in Next.js 15", engine: "all" })
27
+ ```
28
+
29
+ **Engines:**
30
+ - `all` — fan-out to all three in parallel (default, highest confidence)
31
+ - `perplexity` — best for technical Q&A
32
+ - `bing` — best for recent news and Microsoft ecosystem
33
+ - `google` — best for broad coverage
34
+
35
+ ## Requirements
36
+
37
+ - Chrome must be running (or it auto-launches a dedicated instance via `launch.mjs`)
38
+ - The `chrome-cdp` skill must be accessible (same CDP infrastructure as GreedySearch-claude)
39
+
40
+ ## Setup (first time)
41
+
42
+ To pre-launch the dedicated GreedySearch Chrome instance:
43
+
44
+ ```bash
45
+ node ~/.pi/agent/git/GreedySearch-pi/launch.mjs
46
+ ```
47
+
48
+ Stop it when done:
49
+
50
+ ```bash
51
+ node ~/.pi/agent/git/GreedySearch-pi/launch.mjs --kill
52
+ ```
53
+
54
+ ## How It Works
55
+
56
+ - `index.ts` — Pi extension, registers `greedy_search` tool
57
+ - `search.mjs` — CLI runner, spawns extractors in parallel
58
+ - `launch.mjs` — launches dedicated Chrome on port 9223
59
+ - `extractors/` — per-engine CDP scrapers (Perplexity, Bing Copilot, Google AI)