io.github.beycom/onetool-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/integration/tools/test_mem.py: contains a hardcoded OpenAI API key (sk-a…23 (len 28)) — in an example/sample file; verify it is not a real key. tests/unit/core/test_proxy_manager.py: assigns a literal to token (${GI…N} (len 15)) — non-provider secret literal; prefer environment intake. tests/unit/core/test_secrets.py: assigns a literal to API_KEY (${MY…E} (len 18)) — non-provider secret literal; prefer environment intake. tests/unit/core/test_secrets.py: assigns a literal to API_KEY (${NO…R} (len 18)) — non-provider secret literal; prefer environment intake. tests/unit/core/test_secrets.py: assigns a literal to API_KEY (orig…ey (len 12)) — non-provider secret literal; prefer environment intake. tests/unit/core/test_secrets.py: assigns a literal to API_KEY (secr…23 (len 14)) — non-provider secret literal; prefer environment intake. tests/unit/core/test_secrets.py: assigns a literal to API_KEY (from…ts (len 12)) — non-provider secret literal; prefer environment intake. tests/unit/tools/test_mem.py: contains a hardcoded OpenAI API key (sk-a…23 (len 28)) — in an example/sample file; verify it is not a real key. Reads secrets from the environment (e.g. src/bench/run.py) — the recommended intake shape.

• Permission scope

  fail15% of grade

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

  HIGH: tests/unit/tools/test_mem.py: references a sensitive credential path (/.ssh/) — the tool-poisoning exfiltration target. HIGH: shell/exec capability AND network egress together — download-and-execute risk (scripts/release_publish.py: subprocess.* call with shell=True (command-injection prone); src/bench/harness/client.py: outbound HTTP/socket call). HIGH: shell/exec capability AND destructive filesystem writes together (scripts/release_publish.py: subprocess.* call with shell=True (command-injection prone); scripts/release_publish.py: destructive filesystem call (shutil.rmtree/os.remove/etc.)). discloses shell/exec capability: scripts/release_publish.py: subprocess.* call with shell=True (command-injection prone) discloses shell/exec capability: src/ot/executor/linter.py: subprocess.* call discloses shell/exec capability: src/ot/executor/validator.py: subprocess.* call discloses shell/exec capability: src/ot/executor/validator.py: os.system/os.popen/os.exec* call discloses shell/exec capability: src/ot/executor/worker_pool.py: subprocess.* call discloses shell/exec capability: src/ot/utils/truncate.py: subprocess.* call discloses shell/exec capability: src/ot_tools/ripgrep.py: subprocess.* call discloses shell/exec capability: tests/smoke/test_cli_bench.py: subprocess.* call discloses shell/exec capability: tests/smoke/test_cli_onetool.py: subprocess.* call discloses shell/exec capability: tests/unit/core/test_code_validator.py: subprocess.* call discloses shell/exec capability: tests/unit/core/test_code_validator.py: os.system/os.popen/os.exec* call discloses dynamic code execution: src/ot/executor/pack_proxy.py: uses eval/exec/__import__ discloses dynamic code execution: src/ot/executor/runner.py: uses eval/exec/__import__ discloses dynamic code execution: src/ot/executor/validator.py: uses eval/exec/__import__ discloses dynamic code execution: tests/smoke/test_cli_onetool.py: uses eval/exec/__import__ discloses dynamic code execution: tests/unit/core/test_code_validator.py: uses eval/exec/__import__ discloses broad/destructive filesystem access: scripts/release_publish.py: destructive filesystem call (shutil.rmtree/os.remove/etc.) discloses broad/destructive filesystem access: src/bench/run.py: open(…, 'w'/'a'/'x') write-mode file access discloses broad/destructive filesystem access: src/onetool/cli.py: destructive filesystem call (shutil.rmtree/os.remove/etc.) discloses broad/destructive filesystem access: src/ot/meta.py: open(…, 'w'/'a'/'x') write-mode file access discloses broad/destructive filesystem access: src/ot/paths.py: destructive filesystem call (shutil.rmtree/os.remove/etc.) discloses broad/destructive filesystem access: src/ot/logging/config.py: path traversal ('../') combined with file access discloses broad/destructive filesystem access: src/ot/stats/jsonl_writer.py: open(…, 'w'/'a'/'x') write-mode file access discloses broad/destructive filesystem access: src/ot_tools/diagram.py: path traversal ('../') combined with file access discloses broad/destructive filesystem access: src/ot_tools/file.py: open(…, 'w'/'a'/'x') write-mode file access discloses broad/destructive filesystem access: src/ot_tools/file.py: destructive filesystem call (shutil.rmtree/os.remove/etc.) discloses broad/destructive filesystem access: tests/unit/tools/test_mem.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

  unscannable10% of grade

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

  no prior version to diff (first sight of io.github.beycom/[email protected]); version-drift / rug-pull cannot be evaluated. Cold-start risk is covered by point-in-time checks (injection_surface, supply_chain, credential_hygiene), not here.

• 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