Pil MCP server

by pjpoulose·io.github.pjpoulose/pil·v0.1.1·1 stars

No reviews yet

PIL: local-first knowledge base for your Instagram saved posts, with a read-only MCP server.

A89/100grade A

Write a review Install

What users say

No reviews yet

Be the first

Safety scan

A89/100

full report

Adoption

Growing

1 stars34 downloads/wk

Reviews

Write one

Nobody has reviewed Pil yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

Pil tools (5, 1 write)

write = sends, deletes, buys or posts

Read from the package source without running it. The installed server may list more.

get_postwrite action
Return everything PIL knows about one post: metadata, deep-read knowledge (summary, key_points, howto, links), folders, and tags.
library_stats
Totals for the library plus deep-read coverage per field.
list_folders
List the user's saved-post collections (rooms). `item_count` is the count Instagram reports; `mapped_posts` is how many are actually indexed.
list_tags
List tags by usage, most-used first.
search_posts
Search the library. Matches query text against captions and deep-read knowledge (summaries, key points, how-to). Optionally restrict to one folder (collection id or folder name) and/or one tag. Returns up to `limit` matches (1-50) as compact cards with a `has_knowledge` flag.

Public scan report

scanner v0.1.10 · 2026-10-03 · same rubric, same numbers if you re-run it

no findings

Code scan11 source files scanned25/25
–Live reliabilityno gateway calls yet and no remote to proben/a
–Tool poisoningtools not inspected (local package is not executed); not countedn/a
Auth qualitylocal package, no credentials required12/15
Maintenancelast push 18 days ago15/15
Maintainer identityregistry namespace matches repository owner6/10

Overall 89/100. Components that don't apply are left out of the denominator. Any critical finding is an F.Rubric Appeal a finding JSON

How Pil compares

Against every graded MCP server, from our own scans and uptime checks. Updated after each scan run.

Safety score: 89/100; higher than 80% of 48,243 servers
Tools: 5; the typical server has 7
Tools that change things: 20%; typical: 0%. Fewer write tools means less an agent can break
GitHub stars: 1; more than 65% of 28,706 servers
Downloads a week: 34; more than 23% of 12,306 servers with a package

What its tools let an agent do: 4 read · 1 change

Install Pil in Claude Code, Cursor or VS Code

claude mcp add pil -- uvx personal-instagram-library

Add to Cursor

What the publisher says

From the Pil repository's README, as published. We do not edit it. Read it on GitHub

Your Instagram saved posts — deep-read by AI, searchable forever, on your machine.

Get started · How it works · Other AI tools · What's inside

mcp-name: io.github.pjpoulose/pil

### Get PIL in 3 steps

1. Copy-paste this into Muse:

Clone https://github.com/pjpoulose/PIL into your workspace and follow its

SKILL.md to set up my Personal Instagram Library. Work through it step by

step — install, build the library from my saved posts, and hand me the

installable app.

2. When it asks, link your Instagram (one tap in your browser).

3. Download the app file it sends you → unzip → double-click Start PIL → click Install.

That's the whole thing. Your Muse does the setup, the reading, and the building.

Want it on your phone too? Just ask your Muse — it handles that as well.

Prefer to do it yourself? One command sets everything up.

- Mac / Linux: curl -fsSL https://raw.githubusercontent.com/pjpoulose/PIL/master/bootstrap.sh | bash

- Windows (PowerShell): irm https://raw.githubusercontent.com/pjpoulose/PIL/master/bootstrap.ps1 | iex

Then build your library with the commands in Manual setup.

Every day you save posts you'll never find again. PIL (Personal Instagram Library) turns your Instagram saved collection into a private knowledge base on your own machine: every post deep-read by vision AI — narrative summaries, key points, how-to steps, links it discusses, automatic tags — searchable in seconds and queryable live from your AI coding tools.

Code is shared, data stays home. This repo contains only code, the database schema, config examples, and docs. Your saved posts, captions, and account details never leave your computer — ingestion, extraction, search, and the MCP server all run locally. A .gitignore blocks databases, configs, and exports from ever being committed.

Concept mockup with sample data — your library looks like this, with your posts.

Want your Claude, Codex, or Cursor to access this?

Connect the read-only MCP server and your other AI tools can query your library live — always current, nothing to re-upload. Your Muse can wire it up for you. How to connect →

How it works

flowchart LR
    A["Instagram saved posts"] -->|"ingest"| B["Local SQLite on your machine"]
    B -->|"vision AI<br/>deep read"| C["Summaries, key points, how-tos, links, tags"]
    C --> D["Query two ways"]
    D --> E["MCP server — Cursor, Claude Code, Desktop"]
    D --> F["Static JSON export — any AI tool"]

Manual setup (do it yourself)

Shortened. The full README is on GitHub.

Nothing above is checked by us. What we check is on the safety report.

Pil: common questions

Is Pil MCP server safe?: Yes, by our scan: it is graded A (89/100). Read the Pil safety report
How do I install Pil?: It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
Does Pil need an API key?: Not as far as the registry entry and our scan can tell: no credentials are declared or required.
Is Pil maintained?: The last commit was 51 days ago (2026-09-15). The latest release is v0.1.1.
How does Pil compare with other MCP servers?: safety score 89/100 (higher than 80% of 48,243 servers); tools 5 (the typical server has 7); tools that change things 20% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.