Is PMB AI MCP server safe?

Probably. Read the findings first.

C66/100grade C

Use with care. Some checks failed or could not be verified.

What to know before installing

How PMB AI compares

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

Safety score: 66/100; higher than 23% of 48,255 servers
Tools that change things: 4%; typical: 0%. Fewer write tools means less an agent can break

What its tools let an agent do: 63 read · 4 change · 2 run or deploy

scanner v0.1.9 · 2026-09-26 · same rubric, same numbers if you re-run it

1 high1 medium

Code scan172 source files scanned8/25
–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 3 days ago15/15
Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10

highShell command built from a string (injection risk)exec.shell-concat

pmb_ai-1.1.0/src/pmb/cli/commands/config.py: …ad" if os.name == "nt" else "nano") os.system(f'{editor} "{path}"') …

mediumeval / new Function usedexec.eval

pmb_ai-1.1.0/src/pmb/config.py: …fault - and a real-workspace " "eval (June 8 2026) with the default MiniLM emb…

Overall 66/100. Components that don't apply are left out of the denominator. Any critical finding is an F.Rubric Appeal a finding JSON

Deja Vu
deja-vu: local memory over the session histories of forty-three coding agents.
not reviewed
Established
A
Memorix
Local-first project memory with legacy MCP and 2026 discovery compatibility.
not reviewed
Established
B
Nautilus Compass Memory
Local-first long-term memory for AI agents, with byte-recomputable signed verification receipts.
not reviewed
Established
B