Is pinelabs-mcp-troubleshoot safe to install?

SaferSkills independently audited pinelabs-mcp-troubleshoot (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.

CategoryWeightCategory scoreContribution

Securityprompt, exec, net, exfil, eval

35%

100

35.0 pts

Supply chainhash, typosquat, maintainer, lockfile

20%

100

20.0 pts

Maintenancestaleness, pinning, CI

15%

100

15.0 pts

TransparencySKILL.md, perms, README

15%

100

15.0 pts

Communityinstalls, verify, response

15%

100

15.0 pts

Findings & checks · 0 flagged

Securityscore 100 · 0 findings

✓—All security checks passedNo findings in this category for the latest scan.pass

Supply chainscore 100 · 0 findings

✓—All supply chain checks passedNo findings in this category for the latest scan.pass

Maintenancescore 100 · 0 findings

✓—All maintenance checks passedNo findings in this category for the latest scan.pass

Transparencyscore 100 · 0 findings

✓—All transparency checks passedNo findings in this category for the latest scan.pass

Communityscore 100 · 0 findings

✓—All community checks passedNo findings in this category for the latest scan.pass

The primary manifest — the file an agent reads to learn what this artifact does.

.agents/skills/pinelabs-mcp-troubleshoot/SKILL.md · 1 file

.agents/skills/pinelabs-mcp-troubleshoot/SKILL.md6.2 KB · Markdown

name: pinelabs-mcp-troubleshoot

description: "Diagnose and fix Pine Labs MCP server issues. Use when: the server won't start, tools aren't showing up, authentication fails, API calls return errors, the MCP client can't connect, or tests fail. Covers connection problems, credential issues, environment mismatches, timeout errors, token cache problems, and Docker runtime issues."

Pine Labs MCP Troubleshoot

Diagnose and resolve issues with the Pine Labs MCP server running locally.

When to Use

Server won't start or crashes immediately
MCP client shows "server disconnected" or "failed to connect"
Authentication errors (401, invalid credentials, token failures)
Tools don't appear in the MCP client
API calls return unexpected errors
Tests fail after setup
Docker container exits or won't build

Diagnostic Procedure

Follow this sequence — each step narrows the problem category.

Step 1: Verify Python Environment

# Check Python version (need 3.10+)
python3 --version

# Check virtual environment is active
which python
# Should point to .venv/bin/python, NOT system Python

# If not active:
source .venv/bin/activate

# Check dependencies installed
pip list | grep -E "fastmcp|httpx|pydantic"
# All three must be present

Common fix: pip install -r requirements.txt

Step 2: Verify Server Starts

# Start with debug logging
python -m cli.pinelabs_mcp_server.main stdio \
  --client-id <ID> --client-secret <SECRET> \
  --env uat --log-level DEBUG --log-file /tmp/mcp-debug.log

Check the log file for errors:

cat /tmp/mcp-debug.log

Common errors:

Error	Cause	Fix
`ModuleNotFoundError: cli`	Wrong working directory	`cd` to project root
`ModuleNotFoundError: fastmcp`	Dependencies not installed	`pip install -r requirements.txt`
`ImportError`	Python version too old	Upgrade to 3.10+
`Address already in use`	Port conflict (HTTP mode)	Use different `--port` or kill existing process

Step 3: Verify Credentials

# Test authentication directly
python -c "
import asyncio
from pkg.pinelabs.client import PineLabsClient

async def test():
    client = PineLabsClient(
        client_id='<ID>',
        client_secret='<SECRET>',
        base_url='https://pluraluat.v2.pinepg.in/api',
        token_url='https://pluraluat.v2.pinepg.in/api/auth/v1/token',
    )
    token = await client._get_token()
    print(f'Token obtained: {token[:20]}...')
    await client.close()

asyncio.run(test())
"

Common errors:

Error	Cause	Fix
`401 Unauthorized`	Wrong client ID or secret	Verify credentials with Pine Labs dashboard
`ConnectionError`	Network/firewall blocking	Check internet, VPN, proxy settings
`TimeoutError`	API unreachable	Check `PINELABS_BASE_URL` matches environment
`SSL certificate error`	Corporate proxy/MITM	Set `SSL_CERT_FILE` env var or check proxy config

Step 4: Verify MCP Client Config

#### VS Code

Open Command Palette → MCP: List Servers
Check if "pinelabs" appears
If not, verify .vscode/mcp.json exists and has correct syntax
Check Output panel → "MCP" channel for errors
Verify python command resolves to the correct interpreter (venv)

Fix for wrong Python path: Use absolute path in config:

"command": "/path/to/project/.venv/bin/python"

#### Claude Desktop

Check config at ~/Library/Application Support/Claude/claude_desktop_config.json
Verify JSON is valid (no trailing commas)
Restart Claude Desktop after config changes
Check Console.app for claude process errors (macOS)

#### Cursor

Check .cursor/mcp.json in project root
Verify JSON syntax
Restart Cursor after changes
Check Developer Console (Help → Toggle Developer Tools)

Step 5: Verify Tools Are Registered

# Run tests to confirm tool registration works
make test

If only some tools are missing, check:

--toolsets flag — may be limiting which toolsets load
--read-only flag — hides write tools
Check pkg/pinelabs/tools.py for registration order

Step 6: Check Docker Issues

# Verify image builds
docker build -t pinelabs-mcp-server .

# Run with debug logging
docker run --rm \
  -e PINELABS_CLIENT_ID=<ID> \
  -e PINELABS_CLIENT_SECRET=<SECRET> \
  -e PINELABS_ENV=uat \
  -e LOG_LEVEL=DEBUG \
  pinelabs-mcp-server

# Check for permission issues (non-root user)
docker run --rm --entrypoint sh pinelabs-mcp-server -c "whoami && ls -la /app"

Common Docker issues:

Issue	Cause	Fix
Build fails at pip install	Network in build context	Use `--network host` or configure proxy
Container exits immediately	Missing env vars	Pass all required env vars via `-e` or `.env`
Permission denied	Non-root user can't access files	Check file ownership in Dockerfile
Can't connect from host	stdio transport needs special handling	Use `docker compose` from `examples/`

Error Pattern Reference

API Errors (from tool responses)

Response Code	Meaning	Action
`VALIDATION_ERROR`	Invalid input to tool	Check parameter format (IDs, amounts)
`NOT_FOUND`	Resource doesn't exist	Verify ID, check environment (uat vs prod data)
`UNAUTHORIZED`	Token expired or invalid	Restart server to refresh token cache
`RATE_LIMITED`	Too many requests	Wait and retry; reduce request frequency
`INTERNAL_ERROR`	Unexpected server-side error	Check logs, report if persistent

Environment Mismatches

Symptom	Likely cause
Resources exist in dashboard but API returns 404	Wrong `--env` (uat vs prod)
Auth works but no data	Credentials are for different environment
Amounts look wrong	Pine Labs uses paisa (100 paisa = ₹1)

Quick Fixes Checklist

[ ] Python 3.10+ and venv activated?
[ ] pip install -r requirements.txt done?
[ ] Running from project root directory?
[ ] Credentials correct for the target environment?
[ ] Network access to pluraluat.v2.pinepg.in or api.pluralpay.in?
[ ] MCP client config has valid JSON?
[ ] MCP client restarted after config change?
[ ] make test passes?
[ ] Log file shows successful token acquisition?

pinelabs-mcp-troubleshoot— agent skill

Is pinelabs-mcp-troubleshoot safe to install?

Pine Labs MCP Troubleshoot

When to Use

Diagnostic Procedure

Step 1: Verify Python Environment

Step 2: Verify Server Starts

Step 3: Verify Credentials

Step 4: Verify MCP Client Config

Step 5: Verify Tools Are Registered

Step 6: Check Docker Issues

Error Pattern Reference

API Errors (from tool responses)

Environment Mismatches

Quick Fixes Checklist

Audit the pieces. Scan the whole. Decide.