io.github.Krv-Labs/topos

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

  pass25% of grade

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

  repository linkage verified

• 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. extensions/vscode/scripts/check-vsix-size.js) — 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: shell/exec capability AND destructive filesystem writes together (extensions/vscode/scripts/stage-binary.js: imports child_process and calls exec/spawn; extensions/vscode/scripts/stage-binary.js: destructive/writing fs.* call (write/unlink/rm/chmod/rename)). discloses shell/exec capability: extensions/vscode/scripts/stage-binary.js: imports child_process and calls exec/spawn discloses shell/exec capability: extensions/vscode/src/extension.ts: imports child_process discloses shell/exec capability: tests/benchmarks/test_cli_startup.py: subprocess.* call discloses shell/exec capability: tests/cli/test_main.py: subprocess.* call discloses shell/exec capability: tests/cli/test_registration_imports.py: subprocess.* call discloses shell/exec capability: tests/mcp/test_changeset.py: subprocess.* call discloses shell/exec capability: tests/mcp/test_depgraph.py: subprocess.* call discloses shell/exec capability: tests/mcp/test_snapshot.py: subprocess.* call discloses shell/exec capability: tests/packaging/conftest.py: subprocess.* call discloses shell/exec capability: tests/packaging/test_hidden_import_parity.py: subprocess.* call discloses shell/exec capability: tests/test_version.py: subprocess.* call discloses shell/exec capability: tests/utils/test_gitnexus.py: subprocess.* call discloses shell/exec capability: topos/cli/update.py: subprocess.* call discloses shell/exec capability: topos/evaluation/security_guidance.py: subprocess.* call with shell=True (command-injection prone) discloses shell/exec capability: topos/mcp/tools/assess/worktree.py: subprocess.* call discloses shell/exec capability: topos/utils/discovery.py: subprocess.* call discloses shell/exec capability: topos/utils/gitnexus.py: subprocess.* call discloses dynamic code execution: tests/cli/test_quality.py: uses eval/exec/__import__ discloses dynamic code execution: tests/evaluation/test_security_guidance.py: uses eval/exec/__import__ discloses dynamic code execution: tests/evaluation/test_suggestions.py: uses eval/exec/__import__ discloses dynamic code execution: tests/evaluation/test_suppression.py: uses eval/exec/__import__ discloses dynamic code execution: tests/functors/profunctors/test_profunctors.py: uses eval/exec/__import__ discloses dynamic code execution: tests/graphs/cpg/test_cpg.py: uses eval/exec/__import__ discloses dynamic code execution: tests/mcp/test_assess.py: uses eval/exec/__import__ discloses dynamic code execution: tests/mcp/test_evaluate.py: uses eval/exec/__import__ discloses dynamic code execution: tests/mcp/test_inspect.py: uses eval/exec/__import__ discloses dynamic code execution: tests/mcp/test_refactor_targets.py: uses eval/exec/__import__ discloses broad/destructive filesystem access: extensions/vscode/scripts/stage-binary.js: destructive/writing fs.* call (write/unlink/rm/chmod/rename) discloses broad/destructive filesystem access: extensions/vscode/scripts/test-stage-binary.js: destructive/writing fs.* call (write/unlink/rm/chmod/rename) discloses broad/destructive filesystem access: extensions/vscode/src/runtime.ts: destructive/writing fs.* call (write/unlink/rm/chmod/rename) discloses broad/destructive filesystem access: extensions/vscode/src/test/unit/runtime.test.ts: destructive/writing fs.* call (write/unlink/rm/chmod/rename) discloses broad/destructive filesystem access: extensions/vscode/src/test/unit/runtime.test.ts: path traversal ('../') combined with file access discloses broad/destructive filesystem access: tests/benchmarks/test_cli_startup.py: destructive filesystem call (shutil.rmtree/os.remove/etc.) discloses broad/destructive filesystem access: topos/cli/installation.py: destructive filesystem call (shutil.rmtree/os.remove/etc.) discloses broad/destructive filesystem access: topos/mcp/snapshots.py: destructive filesystem call (shutil.rmtree/os.remove/etc.) discloses broad/destructive filesystem access: topos/mcp/tools/docs.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.Krv-Labs/[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