Track flow of untrusted data from sources (HTTP params, env vars, file reads) to dangerous sinks (SQL queries, exec, innerHTML, redirects). Framework-aware: knows Express req.params, Laravel $request->input, Django request.GET, FastAPI Query(), etc. Reports unsanitized flows with CWE IDs and fix ...
AI agents call taint_analysis to retrieve information from Trace without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
taint_analysis is a static security scanning and analysis tool that traces data flow through code. It reads source code, builds a dependency graph, and reports potential vulnerabilities without making any modifications or executing external operations. The 'Read-only' designation and JSON return format confirm this is purely informational analysis with no side effects.
From the tool's definition Tool description explicitly states: 'Track flow of untrusted data from sources...to dangerous sinks' and 'Reports unsanitized flows with CWE IDs and fix suggestions'.
Documented attack patterns abuse exactly the kind of access taint_analysis gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Trace, and nothing reaches the server without passing your rules. This is the rule we recommend for taint_analysis:
{
"version": "1",
"default": "deny",
"tools": {
"taint_analysis": {}
}
}taint_analysis is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Track flow of untrusted data from sources (HTTP params, env vars, file reads) to dangerous sinks (SQL queries, exec, innerHTML, redirects). Framework-aware: knows Express req.params, Laravel $request->input, Django request.GET, FastAPI Query(), etc. Reports unsanitized flows with CWE IDs and fix suggestions. More accurate than pattern-based scanning — traces actual data flow paths. Use for data-flow security analysis. For pattern-based OWASP scanning use scan_security instead. Read-only. Returns JSON: { flows: [{ source, sink, path, sanitized, cwe, suggestion }], total }. It is categorised as a Read tool in the Trace MCP Server, which means it retrieves data without modifying state.
Register the Trace MCP server in PolicyLayer and add a rule for taint_analysis: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Trace. Nothing to install.
taint_analysis is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the taint_analysis rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for taint_analysis. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
taint_analysis is provided by the Trace MCP server (nikolai-vysotskyi/trace-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 178 Trace tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
178 Trace tools catalogued and risk-classified — across an index of 42,500+ MCP servers.