io.github.pvliesdonk/scholar-mcp

What we checked

Six static checks, weighted by risk. Every result reflects only what could be observed in the published package and repository — never intent.

• Injection surface

  unscannable25% of grade

  Tool descriptions/manifest scanned for instruction-injection patterns (imperatives at the model, hidden text, 'ignore previous', data-exfil URLs).

  No tool descriptions, server instructions, prompts, or resources were fetched; nothing to scan for injection.

• Supply chain

  fail25% of grade

  Package provenance: namespace verification, repo linkage, maintainer count, account age, postinstall scripts, typosquat distance.

  Python build script executes at build time: subprocess in build script, network call in build script

• Credential hygiene

  pass15% of grade

  How the server takes secrets (env vs plaintext config vs hardcoded); secrets appearing in tool schemas.

  Reads secrets from the environment (e.g. scripts/check_pins.py) — the recommended intake shape. PASS = no static red flag; static analysis cannot prove the code honors env-based secret handling at runtime.

• Permission scope

  warn15% of grade

  Declared tools vs. breadth (filesystem, network, exec); flags shell-exec and unbounded filesystem access.

  has both shell/exec and network egress — capable of download-and-run, though no such path was observed (scripts/gen_config_surface.py: os.system/os.popen/os.exec* call; docs/javascripts/config-wizard/wizard.js: outbound HTTP/socket call). has both shell/exec and destructive filesystem writes (scripts/gen_config_surface.py: os.system/os.popen/os.exec* call; scripts/migrate_agent_instructions.py: destructive filesystem call (shutil.rmtree/os.remove/etc.)). discloses shell/exec capability: 6 files (scripts/gen_config_surface.py, scripts/migrate_agent_instructions.py, scripts/promote_release_notes.py, +3 more) discloses broad/destructive filesystem access: scripts/migrate_agent_instructions.py: destructive filesystem call (shutil.rmtree/os.remove/etc.) Capability DISCLOSURE, not a verdict: static analysis sees the primitive is present and reachable, not whether its use is attacker-controlled.

• Version behavior

  fail10% of grade

  Diff of tool definitions between versions; new permissions or changed descriptions in a patch release (the postmark-mcp class).

  vs prior 1.10.0 (major bump): new capability primitive(s) in source: fs-write, http-client, subprocess; new hardcoded outbound host/BCC destination(s): ..., 127.0.0.1, api.crossref.org, api.github.com, api.openai.com, api.openalex.org, api.semanticscholar.org, api.unpaywall.org, api.w3.org, arxiv.org, auth.example.com, bad.url, books.google.com, covers.openlibrary.org, datatracker.ietf.org, developers.epo.org, docling, docling.invalid, docs.astral.sh, doi.org, etsi.org, example.com, example.invalid, github, github.com, gofastmcp.com, ieeexplore.ieee.org, litellm, mcp.example.com, objects.githubusercontent.com, obp.example.com, openalex.org, openlibrary.org, ops.epo.org, probe, pvliesdonk.github.io, raw.githubusercontent.com, rfc-editor.org, single-market-economy.ec.europa.eu, src.example.com, test, vlm, vlm.example.com, vlm.invalid, webstore.iec.ch, worldwide.espacenet.com, www.commoncriteriaportal.org, www.epo.org, www.etsi.org, www.googleapis.com, www.iso.org, www.loc.gov, www.ncbi.nlm.nih.gov, www.rfc-editor.org, www.semanticscholar.org, www.w3.org, www.worldcat.org, x.

• Transport config

  warn10% of grade

  Remote servers: TLS and auth mode (none/token/OAuth). Local servers: whether the manifest indicates it phones home.

  manifest/source references tracking behavior — disclosure