MCP server health, State of MCP census
Pdf4vllm
io.github.PyJudge/pdf4vllm
PDF reader for vision LLMs. Auto-detects text corruption and switches to image mode.
Checks
The exact rows akashi check io.github.PyJudge/pdf4vllm prints, from the 2026-07-02 census.
| Check | Status | Detail |
|---|---|---|
| registry status | pass | active |
| server.json valid | pass | validates against its declared JSON Schema |
| repo reachable | pass | exists |
| repo freshness | warn | no push in 170d |
| package pypi | pass | published (1.1.3) |
| at least one live entrypoint | pass | 2 alive |
| license present | warn | none detected |
Evidence
Packages
- pypi pdf4vllm-mcp · published(1.1.3)
Remote endpoints
No remote endpoint declared.
server.json
validagainst its declared schema
One of 14559 servers in the 2026-07-02 State of MCP census: 11582 healthy (79.6%), 2212 degraded (15.2%), 596 dead (4.1%), 169 unknown (1.2%).
Verify it yourself
This page is a static read of one dated probe. Run the same check locally, any time, with no key:
$ akashi check io.github.PyJudge/pdf4vllmInstall Akashi - single Go binary, zero keys, MIT licensed.
Is this your server?
Claiming a server (an auto-refreshing badge and a verified publisher profile) is coming soon. Until then, embed the verified-on-date badge yourself with akashi check io.github.PyJudge/pdf4vllm --badge.
Open data, CC BY 4.0. How this census works - denominator, tiers, and what each check does.