io.github.musharna/data-aggregator-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: 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/eval_multi_query.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

  fail15% of grade

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

  HIGH: tests/test_operate.py: references a sensitive credential path (/.ssh/) — the tool-poisoning exfiltration target. HIGH: shell/exec capability AND network egress together — download-and-execute risk (tests/test_search_cli.py: subprocess.* call; scripts/eval_multi_query.py: outbound HTTP/socket call). HIGH: shell/exec capability AND destructive filesystem writes together (tests/test_search_cli.py: subprocess.* call; src/data_aggregator_mcp/archive.py: open(…, 'w'/'a'/'x') write-mode file access). discloses shell/exec capability: tests/test_search_cli.py: subprocess.* call discloses broad/destructive filesystem access: src/data_aggregator_mcp/archive.py: open(…, 'w'/'a'/'x') write-mode file access discloses broad/destructive filesystem access: src/data_aggregator_mcp/archive.py: path traversal ('../') combined with file access discloses broad/destructive filesystem access: src/data_aggregator_mcp/fetch.py: open(…, 'w'/'a'/'x') write-mode file access discloses broad/destructive filesystem access: tests/test_archive.py: path traversal ('../') combined with file access discloses broad/destructive filesystem access: tests/test_datacite.py: path traversal ('../') combined with file access discloses broad/destructive filesystem access: tests/test_fetch.py: path traversal ('../') combined with file access 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 0.41.0 (patch bump): new capability primitive(s) in source: subprocess; new hardcoded outbound host/BCC destination(s): ..., api, api.cellxgene.cziscience.com, api.crossref.org, api.dandiarchive.org, arcticdata.io, cellxgene.cziscience.com, cn, cn.dataone.org, dandiarchive.org, data.openml.org, data.rcsb.org, datasets-server.huggingface.co, datasets.cellxgene, datasets.cellxgene.cziscience.com, down.test, dx.doi.org, e.org, emb.test, err.test, example.com, example.org, example.test, files.rcsb.org, ftp.ebi.ac.uk, ftp.pride.ebi.ac.uk, h, huggingface.co, llm.test, mlcommons.org, mn.example, ns.dataone.org, opendatacommons.org, openml.org, openneuro.org, orcid.org, purl.org, rest, rest.uniprot.org, schema.org, search, search.rcsb.org, u, up.test, w3id.org, www, www.omicsdi.org, www.openml.org, www.rcsb.org, www.uniprot.org, www.w3.org.

• 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 telemetry behavior — disclosure