io.github.mayim-mayim/ln-church-agent-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

  warn15% of grade

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

  tests/test_p0_2_agent_server_contract.py: assigns a literal to secret (RECE…91 (len 29)) — non-provider secret literal; prefer environment intake. tests/test_p0_3_reaudit_identity_contract.py: assigns a literal to token (0x22…22 (len 42)) — non-provider secret literal; prefer environment intake. tests/test_v1_18_0_network_fetch_policy.py: assigns a literal to token (SIGN…b6 (len 37)) — non-provider secret literal; prefer environment intake. tests/test_v1_18_0_task_v2_transport.py: assigns a literal to secret (secr…ue (len 25)) — non-provider secret literal; prefer environment intake. Reads secrets from the environment (e.g. examples/advanced_receipts_and_policy.py) — the recommended intake shape.

• Permission scope

  warn15% of grade

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

  discloses broad/destructive filesystem access: ln_church_agent/cli.py: destructive filesystem call (shutil.rmtree/os.remove/etc.) discloses broad/destructive filesystem access: ln_church_agent/client.py: destructive filesystem call (shutil.rmtree/os.remove/etc.) discloses broad/destructive filesystem access: ln_church_agent/task_journal.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 1.17.1 (minor bump): new capability primitive(s) in source: fs-write, http-client, subprocess; new hardcoded outbound host/BCC destination(s): 127.0.0.1, 127.0.0.1.nip.io, 168.63.129.16, 169.254.169.254, 192.0.0.192, 2130706433, 8.8.8.8, 93.184.216.34, advisor.test, allowed.com, api.devnet.solana.com, api.example.com, api.external.com, api.getalby.com, api.localhost, api.mainnet-beta.solana.com, api.test, b, blocked.test, buyer.test, controlled, database, dest.test, dummy, evil.com, evil.example, example.com, example.org, hello-world-x402.vercel.app, internal, intranet, kari.mayim-mayim.com, legend.lnbits.com, mainnet.base.org, metadata-target.example, metadata.google.internal, mock, mock-relayer, mock.shrine, one.example, other.example, other.invalid, other.shrine, other.test, polygon-rpc.com, provider.test, proxy.invalid, public.example, public.test, redirect.example, service.internal, source.test, surface.test, target.example, tasks-release.mayim-mayim.com, test.com, trusted-issuer, trusted.com, two.example, unknown.com, user, www.iana.org, xn--bcher-kva.example, your-api.com, your-nwc-bridge.com.

• 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