Tachibot
Multi-model orchestration running OpenAI, Google, Anthropic, xAI, Perplexity, and OpenRouter in parallel for cross-validation
0Tools
13Findings
15Stars
Mar 22, 2026Last Scanned
7 critical · 3 high · 2 medium · 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
Findings13
7critical
3high
2medium
1low
Critical7
criticalC1Command InjectionMCP03-command-injectionAML.T0054
Pattern "`[^`]+`" matched in source_code: "`⚠️ Skipping duplicate tool registration: ${tool.name}`" (at position 11905)
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.
criticalJ1Cross-Agent Configuration PoisoningMCP05-privilege-escalationAML.T0054
Pattern "(\.claude[/\\]|claude[/\\]settings|claude.*config)" matched in source_code: "Claude model: ${config" (at position 34429)
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.
criticalK8Cross-Boundary Credential SharingMCP05-privilege-escalationAML.T0054
Pattern "(return|respond|output|result).*(?:token|credential|api[_\s-]?key|secret|password|bearer)" matched in source_code: "result.distilledContext.tokenEstimate} token" (at position 24763)
Never forward, share, or embed credentials across trust boundaries. Use OAuth token exchange (RFC 8693) to create scoped, delegated tokens instead of passing original credentials. Never include credentials in tool responses. Required by ISO 27001 A.5.17 and OWASP ASI03.
criticalK9Dangerous Post-Install HooksMCP10-supply-chainAML.T0054
Pattern "["'](?:postinstall|preinstall|install)["']\s*:\s*["'][^"']*(?:curl|wget|node\s|python|bash|sh\s|powershell)" matched in source_code: ""postinstall": "bash scripts/install-skills.sh " (at position 39287)
Remove network requests, code execution, and shell commands from install hooks. Post-install scripts should only run build/compile steps (node-gyp, tsc). Use --ignore-scripts flag during CI installations and audit all install hooks before allowing. Required by OWASP ASI04 and CoSAI MCP-T11.
criticalP5Secrets Exposed in Container Build LayersMCP07-insecure-configT1552.001
Pattern "ADD\s+.*(?:\.env|credentials|secret|private[_\-]?key|\.pem|\.p12|\.pfx|\.keystore)\s" matched in source_code: "Add them to .env " (at position 35037)
Never pass secrets via Dockerfile ARG/ENV or COPY. Use Docker BuildKit secrets (--secret flag) for build-time secrets, which are not persisted in image layers. For runtime secrets, use mounted Kubernetes Secrets (not ConfigMaps), environment variables sourced from Secret objects, or external secrets managers (Vault, AWS Secrets Manager, Doppler). Enable encryption at rest for Kubernetes etcd. Use multi-stage builds to prevent secret leakage in intermediate layers.
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 143)
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 "(?:next[_\s]?(?:step|node|agent)|route[_\s]?to|redirect[_\s]?to|skip[_\s]?(?:step|node|validation))" matched in source_code: "NEXT STEP" (at position 16310)
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).
High3
highK1Absent Structured LoggingMCP09-logging-monitoringAML.T0054
Pattern "console\.(log|warn|error)\s*\(.*(?:tool|request|handler|execute|invoke)" matched in source_code: "console.warn(`⚠️ Skipping duplicate tool registration: ${tool" (at position 11892)
Implement structured logging (pino, winston, or equivalent) for all tool call handlers. Every tool invocation should log: timestamp, tool name, caller identity, parameters (sanitized), result status, and duration. Required by ISO 27001 A.8.15, CoSAI MCP-T12, and NIST AI RMF MEASURE 2.6.
highK11Missing Server Integrity VerificationMCP10-supply-chainAML.T0054
Pattern "(connect|load|register|add)[_\s-]?(mcp|server|tool)(?!.*(?:verify|validate|checksum|hash|sign|cert|fingerprint|pin))" matched in source_code: "register tool" (at position 9147)
Implement cryptographic verification for MCP server connections: (1) Pin server TLS certificates or public keys, (2) Verify server tool definition checksums against a known-good manifest, (3) Use package manager integrity checks (npm integrity, pip --require-hashes). The MCP spec recommends but doesn't yet mandate server signing — implement it proactively. Required by ISO 27001 A.8.24 and CoSAI MCP-T6.
highK13Unsanitized Tool OutputMCP02-tool-poisoningAML.T0054
Pattern "(?:query|execute|select|find).*(?:return|respond|result|rows|data)(?!.*(?:sanitize|escape|encode|map|filter|select|pick))" matched in source_code: "Execute the tools above in sequence, then call focus with mode="reflect" to synthesize result" (at position 16322)
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.
Medium2
mediumK20Insufficient Audit Context in LoggingMCP09-logging-monitoringAML.T0054
Pattern "console\.(log|warn|error)\s*\(\s*["'`](?:request|handling|processing|executing|tool|invoke)" matched in source_code: "console.error(`Tool" (at position 34256)
Use structured logging that includes all five ISO 27001 A.8.15 fields: (1) WHO — agent/user identity, (2) WHAT — tool name and operation, (3) WHEN — ISO 8601 timestamp, (4) WHERE — server ID and correlation ID, (5) OUTCOME — success/failure and result summary. Replace console.log with structured loggers (pino, winston). Add correlation IDs for request tracing across multi-agent chains.
mediumK17Missing Timeout or Circuit BreakerMCP07-insecure-configAML.T0054
Pattern "(?:query|execute|find|select|aggregate)\s*\((?!.*(?:timeout|maxTimeMS|statement_timeout|deadline|cancel))" matched in source_code: "Execute(" (at position 10410)
Add timeouts to ALL external calls: HTTP requests (30s), database queries (10s), subprocess execution (60s), and MCP tool calls (30s). Implement circuit breakers that open after N consecutive failures (e.g., opossum, cockatiel). Use AbortSignal for cancellable operations. Required by EU AI Act Art. 15 and OWASP ASI08.
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.