Turn HAR-captured web traffic into callable API tools for AI assistants.
This MCP tool does not require its own standalone secrets and is open-source/auditable, but its core function is to replay or call web APIs extracted from HAR files, which may carry forward captured authentication material and send data to domains present in those HARs. Overall it is best rated as caution, with credential handling being the primary security concern.
The description says it extracts and calls web APIs from HAR files; HARs commonly contain cookies, Authorization headers, or session tokens. Even though the tool itself requires no environment secrets, directly reusing captured auth material from HARs creates a clear credential leakage or misuse risk.
Its stated purpose is to call web APIs extracted from HAR files, so network egress is expected; the destinations are not fixed in the materials and instead depend on the sites/endpoints recorded in the HAR. User prompts or parameters may be sent to third-party services referenced by the HAR, but there is no evidence of extra hidden backends or unrelated endpoints.
The system checks indicate that this tool executes code, and as an MCP server it must run locally and handle invocation logic. Based on the available materials, there is no evidence that it requests system privileges beyond its stated purpose, but local process execution should still be treated with caution.
Its functionality at minimum requires reading HAR files and generating callable API tools from them; HARs often include request bodies, response fragments, URL parameters, and identity-related data. The materials do not show a need for broad filesystem writes or excessive data access, but reading HARs alone can expose sensitive browser traffic data.
On the positive side, the project is open-source under MIT, so the code is auditable, which materially reduces opaque supply-chain risk. Caution remains because it comes from a third-party registry, has very low visible adoption (0 stars), and an unknown maintenance status, so source and dependencies should be reviewed before deployment.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "Clarity MCP Server" yet — see the docs or source repo.
I have a HAR file from an admin system. Extract reusable APIs and generate MCP-ready tool definitions, grouped into login, query, and submission actions.
A structured list of API tools with parameter descriptions, action groups, and tool definitions usable by an AI.
Using this HAR file, identify the requests for searching orders and viewing details, convert them into MCP tools, and explain the required input parameters for each tool.
Order lookup tools and their parameter schemas, ready for AI-driven business data queries.
Analyze all endpoints in this HAR file, deduplicate them, summarize which requests are suitable as MCP tools, and flag any high-risk or session-dependent endpoints.
An API capability assessment with toolable endpoints, risk notes, and recommendations for use.
Parse and analyze HAR files to troubleshoot network traffic with automatic redaction.
Load and analyze HAR files locally with structured summaries for AI assistants.
Reduce LLM token usage by lazy-loading tools and routing repetitive subtasks.
Let AI browse via your real Chrome for extraction and multi-step workflows.
Convert OpenAPI specs into MCP servers so AI agents can call APIs.
Deploy an auth-free web search MCP server for AI tool integration.