Industrial-Grade Post-Quantum Encryption Engine and MCP Server
Maknoon provides 11 cryptographic tools with complete input schemas and descriptions. All tools have verb-noun naming (encrypt_file, decrypt_file, generate_keypair, etc.) and descriptions ranging 60 - 180 chars, meeting baseline expectations. However, output schemas are not documented, LLMs cannot predict return structure or chain tools effectively. Parameter descriptions are present but generic; many lack format constraints, ranges, or validation guidance. Error handling is not visible in tool definitions. Security-sensitive tools (encrypt_file, generate_keypair) accept passphrases as parameters, violating secret-injection patterns. No tool annotations (readOnlyHint, destructiveHint, idempotentHint) are declared, limiting agent reasoning about side effects.
Export cryptographic audit log entries with optional filtering
Verify the hash-chain integrity of the audit log to detect tampering
Decrypt a file encrypted with Maknoon using symmetric or asymmetric post-quantum cryptography
Encrypt a file or directory using symmetric or asymmetric post-quantum cryptography
Split a file into encrypted shards using Shamir's Secret Sharing
Generate a new post-quantum cryptographic keypair (KEM and signature keys)
Generate a cryptographically secure random passphrase
Get metadata about an encrypted file (cipher, profile, compression status)
Output schemas not documented. LLMs cannot predict return structure (e.g., what fields does encrypt_file return? Is there a file_hash, size, or metadata?). This breaks tool chaining and forces agents to guess downstream parameter names.
Passphrases accepted as tool parameters (encrypt_file, decrypt_file, generate_keypair, fragment_file, reassemble_file, sign_data). Secrets in parameters are logged in traces and prompt history. Use server-side secret injection via environment variables or vault.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 53 | 2026-07-28+ | v2 |
Reconstruct a file from encrypted shards using Shamir's Secret Sharing
Sign data or a file using post-quantum signature algorithm
Verify a post-quantum signature on data or a file
No tool annotations (readOnlyHint, destructiveHint, idempotentHint). Agents cannot reason about side effects. encrypt_file and decrypt_file modify the filesystem but lack destructiveHint. audit_verify and verify_signature are read-only but lack readOnlyHint.
Parameter descriptions lack format constraints and validation guidance. 'profile' in encrypt_file and generate_keypair is underdescribed (what are valid values? NIST1, NIST3, NIST5?). 'threshold' in fragment_file lacks range (1 - total_shards?). LLMs cannot self-correct without explicit constraints.
audit_verify has empty input schema (no parameters). Description does not explain what it verifies or what output to expect. Agents cannot determine when to call it or how to interpret results.