Adaptive Memory Graph
Persistent memory plugin for Claude using weighted, interconnected knowledge nodes with time-based decay.
0Tools
8Findings
0Stars
Mar 24, 2026Last Scanned
3 critical · 4 high · 1 low findings detected
Security Category Deep Dive
Prompt Injection
Prompt & context manipulation attacks
69
Maturity
14
Rules
5
Sub-Categories
1
Gaps
64%
Implemented
56
Tests
1
Stories
100%3 rules
Injection via tool descriptions and parameter fields
GAP-001Prompt Injection Coverage GapMissing detection coverage for emerging prompt injection attack variants not addressed by current rules
100%4 rules
Hidden instructions via external content and tool responses
100%2 rules
Context window saturation and prior-approval exploitation
100%3 rules
Payload hiding via invisible chars, base64, schema fields
100%2 rules
Injection via prompt templates and runtime tool output
Findings8
3critical
4high
1low
Critical3
criticalJ1Cross-Agent Configuration PoisoningMCP05-privilege-escalationAML.T0054
Pattern "(\.claude[/\\]|claude[/\\]settings|claude.*config)" matched in source_code: ".claude/" (at position 15937)
MCP servers MUST NOT write to AI agent configuration directories. If configuration management is the server's purpose, require explicit user confirmation before any write and restrict paths to the server's own config namespace. CVE-2025-53773 demonstrated this enables full RCE across agent boundaries.
criticalQ13MCP Bridge Package Supply Chain AttackMCP10-supply-chainAML.T0054
Pattern "(?:mcp|fastmcp|langchain-mcp|llama-index-mcp)(?:>=|~=|==)?(?!\d)" matched in source_code: "MCP" (at position 35)
MCP bridge packages (mcp-remote, mcp-proxy, @modelcontextprotocol/sdk, fastmcp) are high-value supply chain targets — CVE-2025-6514 (CVSS 9.6) in mcp-remote affected 437,000+ installs. Always pin exact versions (no ^ or ~ ranges). Use lockfiles (package-lock.json, pnpm-lock.yaml, uv.lock). Never run `npx mcp-remote` without version pinning. Verify package integrity with `npm audit` or `pip-audit` before deployment. Reference: CVE-2025-6514, OWASP ASI04.
criticalQ9Agentic Workflow DAG ManipulationMCP05-privilege-escalationAML.T0058
Pattern "(?:add|remove|modify|update|insert|delete)[_\s]?(?:node|edge|step|task|stage)" matched in source_code: "remove_node" (at position 10028)
MCP tools must NOT modify agentic workflow graphs, execution order, or routing during runtime. Workflow structure should be immutable once execution begins. If dynamic workflow modification is required, implement: (1) approval gates that require human confirmation before graph mutations, (2) integrity checks that validate the workflow graph against a known-good baseline after each step, (3) audit logging of all graph modifications with rollback capability. Reference: arXiv 2602.19555, Trend Micro subgraph impersonation attack (2026).
High4
highQ14Concurrent MCP Server Race ConditionMCP07-insecure-configT1068
Pattern "(?:read|write|modify|delete).*(?:file|path|directory)(?!.*(?:lock|mutex|semaphore|flock|atomic))" matched in source_code: "read_chat_session(session_path" (at position 18995)
MCP servers sharing filesystem or database backends with other servers must implement proper concurrency controls. Use: (1) file locking (flock/lockfile) for filesystem operations, (2) database transactions for all read-modify-write sequences, (3) atomic file operations (O_EXCL, mkdtemp) instead of check-then-create, (4) lstat() to detect symlinks before following (CVE-2025-53109). Never assume exclusive access to shared resources — other MCP servers may be operating concurrently.
highN1JSON-RPC Batch Request AbuseMCP07-insecure-configAML.T0054
Pattern "for\s+\w+\s+in\s+(?:batch|requests|messages)\s*:(?!.*(?:len\(|max_batch|BATCH_LIMIT|islice))" matched in source_code: "for msg in messages:" (at position 19996)
Enforce a maximum batch size for JSON-RPC requests (recommended: 20-25 requests per batch). Reject batches exceeding the limit with a -32600 (Invalid Request) error. Implement per-batch timeout and memory limits. Reference: JSON-RPC 2.0 Section 6 (Batch) does not mandate limits — servers MUST enforce them.
highK13Unsanitized Tool OutputMCP02-tool-poisoningAML.T0054
Pattern "(?:fetch|axios|requests?\.get|http\.get).*(?:return|respond|result|body|text|data)(?!.*(?:sanitize|escape|encode|strip|validate|parse|extract))" matched in source_code: "Fetch the full content of a memory node when context" (at position 1812)
Sanitize all external data before including in tool responses. Implement output encoding that neutralizes prompt injection patterns. Truncate excessively long content. Validate structure before passing database results. Apply the principle: treat all external data as untrusted, even in tool outputs. Required by CoSAI MCP-T4.
highD1Known CVEs in DependenciesMCP08-dependency-vuln
Dependency "mcp@null" has known CVEs:
Update dependencies to versions that patch known CVEs. Run 'npm audit fix' or 'pip-audit' to identify and resolve vulnerable dependencies.
Low1
lowF4MCP Spec Non-ComplianceMCP07-insecure-config
Server fails MCP spec compliance checks: required:server_name; required:server_version; required:protocol_version; recommended:tool_descriptions; recommended:parameter_descriptions
Follow the MCP specification for server metadata. Include server name, version, and protocol version. Provide descriptions for all tools and parameters.