Recursive Thinking
Token-efficient recursive thinking engine that guides iterative refinement through focused prompts and context compression.
0Tools
5Findings
2Stars
—Downloads
Mar 22, 2026Last Scanned
Findings5
2critical
2high
0medium
1low
0informational
criticalC1Command InjectionMCP03-command-injectionAML.T0054
Pattern "`[^`]+`" matched in source_code: "`Chain-of-Thought recursive thinking engine with self-evaluation.
HOW IT WORKS:
1. Start with action='start' and your task
2. AI receives phase-specific prompts (explore → justify → evaluate → rethink → finalize)
3. AI explains reasoning, rates its solution, and rethinks if needed
4. Continues until production-ready (confidence ≥85%, rating ≥7/10)
KEY FEATURES:
- Chain of Thought: AI explains WHY it chose each solution
- Self-Evaluation: AI rates solutions (1-10 scale, good/bad flag)
- Rethink Loop: Automatically rethinks if rating <7 or issues found
- Token Efficient: Compresses history, phase-focused prompts
PHASES:
1. EXPLORE: Propose solution with reasoning
2. JUSTIFY: Critically examine your reasoning
3. EVALUATE: Rate your solution (1-10, good/bad)
4. RETHINK: Improve based on weaknesses (if needed)
5. FINALIZE: Compile production-ready solution
CONFIGURATION:
{
"maxDepth": 5,
"minConfidence": 0.85,
"minRating": 7, // Minimum score (1-10) to accept
"enableRethinking": true // Enable continuous refinement
}
USAGE:
Call with action='start' to begin, then action='iterate' with your response.`" (at position 945)
Replace exec()/execSync() with execFile() and pass arguments as an array, never as a string. Validate all inputs against an allowlist before use in any shell context. For subprocess.run, always pass a list and shell=False.
criticalQ13MCP Bridge Package Supply Chain AttackMCP10-supply-chainAML.T0054
Pattern "["']@modelcontextprotocol/sdk["']\s*:\s*["'](?:\^|~|\*|latest)" matched in source_code: ""@modelcontextprotocol/sdk": "^" (at position 10287)
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.
highD1Known CVEs in DependenciesMCP08-dependency-vuln
Dependency "@modelcontextprotocol/sdk@1.0.4" has known CVEs:
Update dependencies to versions that patch known CVEs. Run 'npm audit fix' or 'pip-audit' to identify and resolve vulnerable dependencies.
highK16Unbounded Recursion / Missing Depth LimitsMCP07-insecure-configAML.T0054
Pattern "(invoke|call|execute)[_\s-]?(?:tool|agent|self)(?!.*(?:depth|level|limit|max[_\s-]?(?:depth|recursi|iter|call)|count))" matched in source_code: "CallTool" (at position 544)
Add explicit depth/recursion limits to all recursive operations. Use iterative approaches where possible. Set maximum depth for directory walking (max_depth=10), tree traversal (max_level=20), and agent re-invocation (max_calls=5). Implement circuit breakers that halt after N iterations. Required by EU AI Act Art. 15 (robustness) and OWASP ASI08.
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.
Tools
No tools exposed by this server.
Security Category Deep Dive
Sub-Category Tree · Remediation Roadmap · Attack Stories · Compliance Overlay · ATLAS Techniques · Maturity Model
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