# mcp-agent - Doramagic AI Context Pack > Positioning: a pre-install experience and judgment asset. It helps the host AI get off to a good start, but it does not mean the project has already been installed, run, or validated. ## Sufficiency Principle - **Sufficiency over compression**: The AI Context Pack should be sufficient for the host AI to understand the project's value, capability boundaries, entrypoints, risks, and evidence sources before starting work; it may be layered, but it does not aim for the shortest possible summary. - **Compression policy**: Compress only noise and duplication, never context that affects judgment or the quality of the work. ## How the Host AI Should Use This You are reading the AI Context Pack that Doramagic compiled for mcp-agent. Treat it as pre-work context: help the user understand who it fits, what it can do, how to start, what must be verified after install, and where the risks are. Do not claim that you have already installed, run, or executed the target project. ## Claim Consumption Rules - **Fact source**: Repo Evidence + Claim/Evidence Graph; the Human Wiki only supplies salience, terminology, and narrative structure. - **Minimum status for a fact**: `supported` - `supported`: May be used as a project fact, but the answer must cite the claim_id and evidence path. - `weak`: Usable only as a low-confidence lead; the user must be asked to keep verifying. - `inferred`: Usable only for risk notes or open questions; must not be packaged as a project fact. - `unverified`: Must not be used as fact; state clearly that evidence is insufficient. - `contradicted`: Must show the conflicting sources and must not force a single version on the user's behalf. ## Who It Fits Best - **Developers already using host AIs such as Claude/Codex/Cursor/Gemini**: The README or plugin config mentions multiple host AIs. Evidence: `README.md` Claim: `clm_0002` supported 0.86 ## What It Can Do - **Command-Line Startup or Install Flow** (Verify after install): The project documentation contains runnable commands; real use requires running them in a local or host environment. Evidence: `README.md` Claim: `clm_0001` supported 0.86 ## How to Start - `pip install mcp-agent` Evidence: `README.md` Claim: `clm_0003` supported 0.86 ## Continue-or-Stop Decision Card - **Current recommendation**: Needs admin / security approval - **Why**: Continuing may involve secrets, accounts, external services, or sensitive context; get admin or security approval first. ### 30-Second Read - **What to do now**: Needs admin / security approval - **Minimum safe next step**: Run Prompt Preview first; if credentials or an enterprise environment are involved, get approval before trialing - **Do not trust yet**: Role quality and task fit cannot be trusted directly. - **Continuing will touch**: Role selection bias, Command execution, Local environment or project files ### What You Can Trust Now - **Target-audience signal: Developers already using host AIs such as Claude/Codex/Cursor/Gemini** (supported): Backed by a supported claim or project evidence, but that still is not the same as real install results. Evidence: `README.md` Claim: `clm_0002` supported 0.86 - **Capability exists: Command-Line Startup or Install Flow** (supported): You can trust that the project contains signals of this capability; whether it fits your specific task still needs trial or after-install verification. Evidence: `README.md` Claim: `clm_0001` supported 0.86 - **There are Quick Start / install-command signals** (supported): You can trust that the docs mention a startup or install entrypoint; do not run it directly in your primary environment because of that. Evidence: `README.md` Claim: `clm_0003` supported 0.86 ### What You Cannot Trust Yet - **Role quality and task fit cannot be trusted directly.** (unverified): A role library proves there are many roles; it does not prove each one fits your specific task or that a role produces high-quality results. - **Do not treat role copy as real execution capability.** (unverified): Before install you can only judge whether the role description and task profile match; you cannot prove it can complete the task inside the host AI. - **Real output quality cannot be trusted before install.** (unverified): Prompt Preview can only show how it guides you; it cannot prove result quality in the real project. - **Host AI version compatibility cannot be trusted before install.** (unverified): Host loading rules and version differences across Claude, Cursor, Codex, Gemini, and others must be verified in a real environment. - **That it will not pollute your existing host AI's behavior cannot be trusted directly.** (inferred): Skill, plugin, and AGENTS/CLAUDE/GEMINI instructions may change the host AI's default behavior. - **Safe rollback cannot be assumed by default.** (unverified): Unless the project clearly provides uninstall and recovery instructions, verify in an isolated environment first. - **After a real install, is it compatible with the user's current host AI version?** (unverified): Compatibility can only be verified in the actual host environment. - **Does the project's output quality meet the user's specific task?** (unverified): The pre-install preview can only show flow and boundaries; it cannot replace real evaluation. ### What Continuing Will Touch - **Role selection bias**: The user's judgment about which expert role should handle the task. Why: Picking the wrong role makes the AI answer from the wrong expert perspective, wasting time or misleading decisions. - **Command execution**: Package managers, network downloads, the local plugin directory, project config, or the user's home directory. Why: Running the very first command can already change your environment; decide whether it is worth running first. Evidence: `README.md` - **Local environment or project files**: Install results, plugin caches, project config, or local dependency directories. Why: The write scope and rollback path cannot be proven before install and need isolated verification. Evidence: `README.md` - **Environment variables / API keys**: Project entry docs explicitly showing API key, token, secret, or account credential configuration. Why: If a real install needs credentials, use test credentials first and go through a permission/compliance review. Evidence: `README.md`, `examples/basic/oauth_basic_agent/README.md`, `examples/oauth/interactive_tool/README.md`, `examples/oauth/interactive_tool/server.py` - **Host AI context**: The AI Context Pack, Prompt Preview, Skill routing, risk rules, and project facts. Why: Importing context affects the host AI's later judgment, so avoid packaging unverified items as facts. ### Minimum Safe Next Steps - **Run Prompt Preview first**: Use an interactive trial to verify the task profile and role match first; do not import the whole role library up front. (applies when: Applies to any project, especially when output quality is unknown.) - **Trial-install only in an isolated directory or a test account**: Avoid letting install commands pollute your primary host AI, real projects, or home directory. (applies when: When there are signals of command execution, plugin config, or local writes.) - **Do not use real production credentials**: Once an environment variable / API key enters the host or toolchain, it can create account and compliance risk. (applies when: When environment signals like API, TOKEN, KEY, or SECRET appear.) - **After install, verify just one minimal task**: Verify loading, compatibility, output quality, and rollback first, then decide whether to use it deeply. (applies when: When moving from a trial into a real workflow.) ### Exit Plan - **Preserve the pre-install state**: Record the original host config and project state so you can later judge whether it is recoverable. - **Keep a record of the original role selection**: If output goes off-topic, you can return to the task-profiling stage and reselect a role instead of pushing on with the wrong one. - **Record the install commands and written paths**: Without clear uninstall instructions, you at least need to know which directories or configs to clean up manually. - **Be ready to revoke test API keys or tokens**: If test credentials leak or are misused, you can cut losses quickly. - **If there is no rollback path, do not enter your primary environment**: No rollback is a blocker before continuing; do not proceed on trust or luck. ## What Can Only Be Previewed - Explain who the project fits and what it can do - Demonstrate a typical conversation flow based on project docs - Help the user decide whether it is worth installing or researching further ## What Must Be Verified After Install - Actually installing the Skill, plugin, or CLI - Running scripts, modifying local files, or accessing external services - Verifying real output quality, performance, and compatibility ## Boundary & Risk Decision Card - **Mistaking the pre-install preview for a real run**: The user may overestimate how much configuration, permission, and compatibility verification the project has already done. Mitigation: Clearly separate prompt_preview_can_do from runtime_required. Claim: `clm_0004` inferred 0.45 - **Command execution will modify the local environment**: Install commands may write to the user's home directory, the host plugin directory, or project configuration. Mitigation: Run in an isolated environment or a test account first. Evidence: `README.md` Claim: `clm_0005` supported 0.86 - **To confirm**: After a real install, is it compatible with the user's current host AI version?. Why: Compatibility can only be verified in the actual host environment. - **To confirm**: Does the project's output quality meet the user's specific task?. Why: The pre-install preview can only show flow and boundaries; it cannot replace real evaluation. - **To confirm**: Do the install commands require network access, permissions, or global writes?. Why: This affects install risk in both enterprise and personal environments. ## Pre-Work Working Context ### Loading Order - First read how_to_use.host_ai_instruction to establish the boundaries of this pre-install judgment asset. - Read claim_graph_summary to confirm facts come from the Claim/Evidence Graph, not the Human Wiki narrative. - Then read intended_users, capabilities, and quick_start_candidates to judge whether the user is a match. - When you need to carry out a concrete task, check role_skill_index first, then evidence_index. - For real install, file modification, network access, performance, or compatibility questions, turn to risk_card and boundaries.runtime_required. ### Task Routes - **Command-Line Startup or Install Flow**: State that this is an after-install capability first, then give a pre-install checklist. Boundary: Must be verified after a real install or run. Evidence: `README.md` Claim: `clm_0001` supported 0.86 ### Context Scale - Total files: 965 - Important-file coverage: 40/965 - Evidence index entries: 80 - Role / Skill entries: 80 ### Handling Insufficient Evidence - **missing_evidence**: State that evidence is insufficient and ask the user for the target file, a README section, or after-install verification records; do not fill in facts. - **out_of_scope_request**: State that the task is beyond the current AI Context Pack's evidence scope and suggest the user check the Human Manual or verify after a real install. - **runtime_request**: Provide a pre-install checklist and command sources, but do not run commands for the user or claim they have been run. - **source_conflict**: Show the conflicting sources side by side, mark them as unverified, and do not force a single version. ## Prompt Recipes ### Fit assessment - Goal: Judge whether this project fits the user's current task. - Expected output: A fit conclusion, key reasons, evidence citations, what can be previewed before install, what must be verified after install, and a next-step recommendation. ```text Based on the AI Context Pack for mcp-agent, ask me 3 necessary questions first, then judge whether it fits my task. The answer must cover: who it fits, what it can do, what it cannot do, whether it is worth installing, and where the evidence comes from. Every project fact must cite evidence_refs, source_paths, or a claim_id. ``` ### Pre-install experience - Goal: Let the user feel the core workflow before installing, while avoiding packaging the preview as real capability or a marketing promise. - Expected output: An experience script with boundary labels, an after-install verification checklist, and a cautious recommendation; with no real-run promises or strong marketing language. ```text Treat mcp-agent as a pre-install experience asset, not an already-installed tool or a real runtime environment. Output exactly four parts: 1. Ask me 3 necessary questions first. 2. Give an "experience script": use the three labels [Previewable before install], [Must verify after install], and [Insufficient evidence] to show how it might guide the workflow. 3. Give an after-install verification checklist: list which capabilities can only be confirmed after a real install, real host loading, and a real project run. 4. Give a cautious recommendation: only "worth researching/trialing further", "add information before deciding", or "not recommended to continue"; do not endorse the project. Hard boundaries: - Do not claim you have installed, run, executed tests, modified files, or produced real results. - Do not write promise-like phrasing such as "auto-adapts", "guarantees passing", "perfect fit", or "strongly recommend installing". - If you describe how it works after install, you must use a conditional such as "if installed successfully and the host loads the Skill correctly, it might...". - The experience script may only be written as "example lines / hypothetical flow": use "might ask / might suggest / might show", not "has written, has generated, has passed, is running, is generating". - Prompt Preview does not hand out install commands; if the user is ready to trial, only prompt them to read Quick Start and the Risk Card first and to verify in an isolated environment. - Every project fact must come from a supported claim, evidence_refs, or source_paths; inferred/unverified items can only be risks or open questions. ``` ### Role / Skill selection - Goal: Pick the best-matching asset from the project's roles or Skills. - Expected output: A list of candidate roles or Skills, each with an applicable scenario, evidence paths, risk boundary, and whether after-install verification is needed. ```text Read role_skill_index and recommend 3-5 of the most relevant roles or Skills for my target task. For each recommendation, state the applicable scenario, likely output, risk boundary, and evidence_refs. ``` ### Risk pre-check - Goal: Identify environment, permission, rule-conflict, and quality risks before installing or adopting. - Expected output: A checklist of environment, permission, dependency, license, host-conflict, quality risk, and unknown items. ```text Based on risk_card, boundaries, and quick_start_candidates, give me a pre-install risk pre-check list. Do not run commands for me; only explain what I should check, why, and what impact a failure would have. ``` ### Host AI kickoff instruction - Goal: Turn the project context into a host AI instruction for the start of a conversation. - Expected output: A pre-work instruction with clear boundaries and clear evidence citations, suitable to copy to a host AI. ```text Based on the AI Context Pack for mcp-agent, generate a pre-work instruction I can paste to my host AI. This instruction must obey not_runtime=true and must not claim the project has been installed, run, or produced real results. ``` ## Role / Skill Index - Indexed 80 role / Skill / project-doc entries. - **mcp-agent Documentation** (project_doc): Install the Mintlify CLI https://www.npmjs.com/package/mintlify to preview the documentation changes locally. To install, use the following command Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `docs/README.md` - **Overview** (project_doc): Build effective agents with Model Context Protocol using simple, composable patterns. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `README.md` - **Agent Factory** (project_doc): This folder shows how to define agents and compose powerful LLM workflows using the helpers in mcp agent.workflows.factory https://github.com/lastmile-ai/mcp-agent/blob/main/src/mcp agent/workflows/factory.py . Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/basic/agent_factory/README.md` - **MCP Functions Agent Example** (project_doc): This example shows a "math" Agent using manually-defined functions to compute simple math results for a user request. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/basic/functions/README.md` - **Basic MCP Agent example** (project_doc): This MCP Agent app shows a "finder" Agent which has access to the fetch https://github.com/modelcontextprotocol/servers/tree/main/src/fetch and filesystem https://github.com/modelcontextprotocol/servers/tree/main/src/filesystem MCP servers. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/basic/mcp_basic_agent/README.md` - **Simplest Usage of MCP Agent - Hello World!** (project_doc): Simplest Usage of MCP Agent - Hello World! Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/basic/mcp_hello_world/README.md` - **LLM Selector example** (project_doc): This example shows using MCP's ModelPreferences type to select a model LLM based on speed, cost and intelligence priorities. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/basic/mcp_model_selector/README.md` - **MCP aggregator example** (project_doc): This example shows connecting to multiple MCP servers via the MCPAggregator interface. An MCP aggregator will combine multiple MCP servers into a single interface allowing users to bypass limitations around the number of MCP servers in use. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/basic/mcp_server_aggregator/README.md` - **MCP Tool Filter Example** (project_doc): This example demonstrates a non-invasive approach to filtering MCP tools without modifying any core mcp-agent code. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/basic/mcp_tool_filter/README.md` - **OAuth Basic MCP Agent example client-only loopback** (project_doc): OAuth Basic MCP Agent example client-only loopback Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/basic/oauth_basic_agent/README.md` - **Streaming Demo** (project_doc): This example demonstrates the real-time streaming capabilities of mcp-agent, showing how to stream LLM responses as they're generated. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/basic/streaming_demo/README.md` - **Token Counter Example** (project_doc): This example demonstrates the MCP Agent's token counting capabilities with custom monitoring and real-time tracking. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/basic/token_counter/README.md` - **mcp-agent Cloud Examples** (project_doc): The examples in this directory are all ready to be deployed to mcp-agent cloud, where they can then be integrated with any MCP client via SSE. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/cloud/README.md` - **Cloud Agent Factory Temporal + Custom Workflow Tasks** (project_doc): Cloud Agent Factory Temporal + Custom Workflow Tasks Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/cloud/agent_factory/README.md` - **ChatGPT App Example** (project_doc): This example demonstrates how to create an MCP Agent application with interactive UI widgets for OpenAI's ChatGPT Apps platform. It shows how to build a coin-flip widget that renders interactive UI components directly in the ChatGPT interface. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/cloud/chatgpt_apps/basic_app/README.md` - **Setup** (project_doc): A basic coin flip component initialized with create-react-app. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/cloud/chatgpt_apps/basic_app/web/README.md` - **Timer App - ChatGPT App Example** (project_doc): ! timer-app https://github.com/user-attachments/assets/7a526501-84c8-4ef5-b784-4b3948790db2 Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/cloud/chatgpt_apps/timer/README.md` - **Setup** (project_doc): A basic coin flip component initialized with create-react-app. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/cloud/chatgpt_apps/timer/web/README.md` - **Hello World Example** (project_doc): This example shows a very basic app with a hello world tool call. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/cloud/hello_world/README.md` - **MCP Server Example** (project_doc): This example is an mcp-agent application that showcases how mcp-agent supports the following MCP primitives: Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/cloud/mcp/README.md` - **Observability Example OpenTelemetry + Langfuse** (project_doc): Observability Example OpenTelemetry + Langfuse Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/cloud/observability/README.md` - **MCP Agent Server Example Temporal** (project_doc): This example demonstrates how to create an MCP Agent Server with durable execution using Temporal https://temporal.io/ . It shows how to build, run, deploy and connect to an MCP server which leverages Temporal workflows for execution. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/cloud/temporal/README.md` - **CrewAI Tools Integration Example** (project_doc): This example demonstrates how to integrate CrewAI tools into MCP Agent workflows. It shows how to use CrewAI's SerperDevTool for web search and FileWriterTool for file operations within an MCP Agent. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/crewai/README.md` - **Human interactions in Temporal** (project_doc): This example demonstrates how to implement human interactions in an MCP running as a Temporal workflow. Human input can be used for approvals or data entry. In this case, we ask a human to provide their name, so we can create a personalised greeting. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/human_input/temporal/README.md` - **Third-Party Tools Integration Example** (project_doc): Third-Party Tools Integration Example Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/langchain/README.md` - **LM Studio Basic Agent Example** (project_doc): This example demonstrates using LM Studio with mcp-agent to run local LLMs with full tool calling and structured output support. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/lm_studio/README.md` - **Elicitation Example** (project_doc): This MCP Agent app shows an Agent which has access to a "Booking System" MCP server. This example highlights the elicitation feature, where a tool can pause its execution to ask the user for additional information or confirmation before proceeding. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/mcp/mcp_elicitation/README.md` - **Deploying the elicitation example to the cloud** (project_doc): Deploying the elicitation example to the cloud Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/mcp/mcp_elicitation/cloud/README.md` - **MCP Primitives Example: Using Resources and Prompts** (project_doc): MCP Primitives Example: Using Resources and Prompts Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/mcp/mcp_prompts_and_resources/README.md` - **SSE example** (project_doc): This example shows how to use an SSE server with mcp-agent. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/mcp/mcp_sse/README.md` - **MCP Agent example** (project_doc): This example shows a basic agent that can connect to an MCP server over SSE with auth headers. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/mcp/mcp_sse_with_headers/README.md` - **MCP Streamable HTTP example** (project_doc): This example shows mcp-agent usage with a Streamable HTTP server using the example server https://github.com/modelcontextprotocol/python-sdk/tree/main/examples/servers/simple-streamablehttp-stateless in the mcp-python repo . Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/mcp/mcp_streamable_http/README.md` - **MCP Websocket example** (project_doc): This example shows a basic agent that can connect to an MCP server over websockets Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/mcp/mcp_websockets/README.md` - **MCP Agent Server Examples** (project_doc): This directory contains examples of exposing MCP Agent workflows as MCP servers. It demonstrates how to build, launch, and interact with agent-powered MCP servers in different execution environments. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/mcp_agent_server/README.md` - **MCP Agent Server Example Asyncio** (project_doc): This example is an mcp-agent application that is exposed as an MCP server, aka the "MCP Agent Server". Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/mcp_agent_server/asyncio/README.md` - **Context Isolation Demo** (project_doc): This example shows how per-request context scoping prevents logs and notifications from bleeding between concurrent MCP clients. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/mcp_agent_server/context_isolation/README.md` - **MCP Agent Server Example Temporal** (project_doc): This example demonstrates how to create an MCP Agent Server with durable execution using Temporal https://temporal.io/ . It shows how to build, run, and connect to an MCP server that uses Temporal as the execution engine. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/mcp_agent_server/temporal/README.md` - **MCP Azure Agent Example - "Finder" Agent** (project_doc): MCP Azure Agent Example - "Finder" Agent Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/model_providers/mcp_basic_azure_agent/README.md` - **MCP Bedrock Agent Example - "Finder" Agent** (project_doc): MCP Bedrock Agent Example - "Finder" Agent Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/model_providers/mcp_basic_bedrock_agent/README.md` - **MCP Google Agent Example - "Finder" Agent** (project_doc): MCP Google Agent Example - "Finder" Agent Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/model_providers/mcp_basic_google_agent/README.md` - **MCP Ollama Agent example** (project_doc): This example shows a "finder" Agent using llama models to access the 'fetch' and 'filesystem' MCP servers. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/model_providers/mcp_basic_ollama_agent/README.md` - **OAuth Examples** (project_doc): Two complementary scenarios demonstrate how OAuth integrates with MCP: Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/oauth/README.md` - **OAuth Interactive Tool Example** (project_doc): This example shows the end-to-end OAuth authorization code flow for a simple synchronous MCP tool. The MCP server exposes a github org search tool that calls the GitHub MCP server. When the tool is invoked without a cached token, the server issues an auth/request message and the client opens the browser so you can complete the GitHub sign-in. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/oauth/interactive_tool/README.md` - **Workflow Pre-Authorize Example** (project_doc): This example shows how to seed OAuth credentials for asynchronous workflows. The client calls the workflows-store-credentials tool to cache a token for a specific workflow before the workflow runs. Once the token is saved, the workflow can access the downstream MCP server without further user interaction. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/oauth/pre_authorize/README.md` - **OAuth protected resource example** (project_doc): This example shows how to integrate OAuth2 authentication to protect your MCP. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/oauth/protected_by_oauth/README.md` - **Temporal Workflow Examples** (project_doc): This collection of examples demonstrates how to use Temporal https://temporal.io/ as the execution engine for MCP Agent workflows. Temporal is a microservice orchestration platform that helps developers build and operate reliable applications at scale. These examples showcase various workflow patterns and use cases. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/temporal/README.md` - **MCP Agent example** (project_doc): This example shows tracing integration in a basic "finder" Agent which has access to the 'fetch' and 'filesystem' MCP servers. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/tracing/agent/README.md` - **Langfuse Trace Exporter Example** (project_doc): This example shows how to configure a Langfuse OTLP trace exporter for use in mcp-agent by adding a typed OTLP exporter with the expected endpoint and headers. Following information from https://langfuse.com/integrations/native/opentelemetry Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/tracing/langfuse/README.md` - **MCP Agent example** (project_doc): This example shows tracing integration for AugmentedLLMs. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/tracing/llm/README.md` - **SSE example** (project_doc): This example shows distributed tracing between a client and an SSE server. mcp-agent automatically propagates trace context in the client requests to the server; the server should be instrumented with opentelemetry and have MCPInstrumentor auto-instrumentation configured from openinference-instrumentation-mcp . Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/tracing/mcp/README.md` - **Temporal Tracing Example** (project_doc): This example demonstrates how to use Temporal https://temporal.io/ as the execution engine for MCP Agent workflows, with OpenTelemetry tracing enabled. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/tracing/temporal/README.md` - **FastAPI WebSocket Example with MCP Agent** (project_doc): FastAPI WebSocket Example with MCP Agent Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/usecases/fastapi_websocket/README.md` - **marimo MCP Agent example** (project_doc): This example marimo https://github.com/marimo-team/marimo notebook shows a "finder" Agent which has access to the 'fetch' and 'filesystem' MCP servers. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/usecases/marimo_mcp_basic_agent/README.md` - **🌐 Browser Console Agent Example** (project_doc): A command-line application that lets you interact with websites using natural language through the Model Context Protocol MCP with the use of the Puppeteer MCP server https://github.com/modelcontextprotocol/servers/tree/main/src/puppeteer . Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/usecases/mcp_browser_agent/README.md` - **MCP Financial Analyzer with Google Search** (project_doc): MCP Financial Analyzer with Google Search Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/usecases/mcp_financial_analyzer/README.md` - **GitHub PRs to Slack Summary Agent** (project_doc): This application creates an MCP Agent that monitors GitHub pull requests and submits prioritized summaries to Slack. The agent uses a LLM to analyze PR information, prioritize issues, and create informative summaries. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/usecases/mcp_github_to_slack_agent/README.md` - **Instagram Gift Advisor** (project_doc): An MCP Agent that analyzes Instagram profiles to generate personalized gift recommendations with real Amazon product links. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/usecases/mcp_instagram_gift_advisor/README.md` - **MCP Marketing Content Agent** (project_doc): This example demonstrates a marketing content creation agent that learns your brand voice and generates platform-optimized content using an evaluation-driven approach with persistent memory for continuous improvement. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/usecases/mcp_marketing_assistant_agent/README.md` - **LinkedIn Candidate Search & CSV Export Tool** (project_doc): LinkedIn Candidate Search & CSV Export Tool Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/usecases/mcp_playwright_agent/README.md` - **MCP Research & Analysis Agent Framework** (project_doc): MCP Research & Analysis Agent Framework Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/usecases/mcp_realtor_agent/README.md` - **MCP Researcher example** (project_doc): This example shows a research assistant agent which has access to internet search via 'brave' https://github.com/modelcontextprotocol/servers/tree/main/src/brave-search , website fetch https://github.com/modelcontextprotocol/servers/tree/main/src/fetch , a python interpreter, and the filesystem https://github.com/modelcontextprotocol/servers/tree/main/src/filesystem . Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/usecases/mcp_researcher/README.md` - **MCP Supabase Migration Agent with GitHub Integration** (project_doc): MCP Supabase Migration Agent with GitHub Integration Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/usecases/mcp_supabase_migration_agent/README.md` - **Reliable Conversation Manager RCM** (project_doc): Implementation of research findings from "LLMs Get Lost in Multi-Turn Conversation" https://arxiv.org/abs/2505.06120 using mcp-agent framework. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/usecases/reliable_conversation/README.md` - **Streamlit MCP Agent example** (project_doc): This Streamlit example shows a "finder" Agent which has access to the 'fetch' and 'filesystem' MCP servers. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/usecases/streamlit_mcp_basic_agent/README.md` - **Streamlit MCP RAG Agent example** (project_doc): This Streamlit example shows a RAG Agent that is able to augment its responses using data from Qdrant vector database. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/usecases/streamlit_mcp_rag_agent/README.md` - **Deep Orchestrator Workflow Example** (project_doc): This example demonstrates the Deep Orchestrator workflow, an adaptive multi-agent system that dynamically plans, executes, and learns from complex tasks. Unlike the standard orchestrator, it features persistent memory, knowledge extraction, budget management, and intelligent replanning capabilities. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/workflows/workflow_deep_orchestrator/README.md` - **Evaluator-Optimizer Workflow Example** (project_doc): Evaluator-Optimizer Workflow Example Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/workflows/workflow_evaluator_optimizer/README.md` - **MCP Agent Intent Classification Workflow example** (project_doc): MCP Agent Intent Classification Workflow example Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/workflows/workflow_intent_classifier/README.md` - **Orchestrator workflow example** (project_doc): This example shows an Orchestrator workflow which dynamically plans across a number of agents to accomplish a multi-step task. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/workflows/workflow_orchestrator_worker/README.md` - **Parallel Workflow example** (project_doc): This example shows a short story grading example. The MCP app runs the proofreader, fact checker, and style enforcer agents in parallel fanning out the calls , then aggregates it together with a grader agent fanning in the results . Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/workflows/workflow_parallel/README.md` - **Workflow Router example** (project_doc): This example shows an LLM-based routing to the top k most relevant categories, which can be an Agent, an MCP server, or a function. The example routes between the functions: print to console , print hello world ; the agents: finder agent , writer agent , reasoning agent . Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/workflows/workflow_router/README.md` - **MCP Swarm Agent** (project_doc): mcp-agent implements OpenAI's Swarm pattern https://github.com/openai/swarm for multi-agent workflows, but in a way that can be used with any model provider. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `examples/workflows/workflow_swarm/README.md` - **MCP Agent Cloud SDK** (project_doc): The MCP Agent Cloud SDK provides a command-line tool and Python library for deploying and managing MCP Agent configurations, with integrated secrets handling. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `src/mcp_agent/cli/README.md` - **Cloud Agent Factory Temporal + Custom Workflow Tasks** (project_doc): Cloud Agent Factory Temporal + Custom Workflow Tasks Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `src/mcp_agent/data/examples/cloud/agent_factory/README.md` - **ChatGPT App Example** (project_doc): This example demonstrates how to create an MCP Agent application with interactive UI widgets for OpenAI's ChatGPT Apps platform. It shows how to build a coin-flip widget that renders interactive UI components directly in the ChatGPT interface. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `src/mcp_agent/data/examples/cloud/chatgpt_app/README.md` - **Setup** (project_doc): A basic coin flip component initialized with create-react-app. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `src/mcp_agent/data/examples/cloud/chatgpt_app/web/README.md` - **Hello World Example** (project_doc): This example shows a very basic app with a hello world tool call. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `src/mcp_agent/data/examples/cloud/hello_world/README.md` - **MCP Server Example** (project_doc): This example is an mcp-agent application that showcases how mcp-agent supports the following MCP primitives: Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `src/mcp_agent/data/examples/cloud/mcp/README.md` - **MCP Agent Server Example Temporal** (project_doc): This example demonstrates how to create an MCP Agent Server with durable execution using Temporal https://temporal.io/ . It shows how to build, run, deploy and connect to an MCP server which leverages Temporal workflows for execution. Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `src/mcp_agent/data/examples/cloud/temporal/README.md` - **MCP Agent Server Example Asyncio** (project_doc): This example is an mcp-agent application that is exposed as an MCP server, aka the "MCP Agent Server". Activation hint: Reference this when the user needs to understand the project's structure, install path, or boundaries. Evidence: `src/mcp_agent/data/examples/mcp_agent_server/asyncio/README.md` ## Evidence Index - Indexed 80 evidence entries. - **mcp-agent Documentation** (documentation): Install the Mintlify CLI https://www.npmjs.com/package/mintlify to preview the documentation changes locally. To install, use the following command Evidence: `docs/README.md` - **Overview** (documentation): Build effective agents with Model Context Protocol using simple, composable patterns. Evidence: `README.md` - **Agent Factory** (documentation): This folder shows how to define agents and compose powerful LLM workflows using the helpers in mcp agent.workflows.factory https://github.com/lastmile-ai/mcp-agent/blob/main/src/mcp agent/workflows/factory.py . Evidence: `examples/basic/agent_factory/README.md` - **MCP Functions Agent Example** (documentation): This example shows a "math" Agent using manually-defined functions to compute simple math results for a user request. Evidence: `examples/basic/functions/README.md` - **Basic MCP Agent example** (documentation): This MCP Agent app shows a "finder" Agent which has access to the fetch https://github.com/modelcontextprotocol/servers/tree/main/src/fetch and filesystem https://github.com/modelcontextprotocol/servers/tree/main/src/filesystem MCP servers. Evidence: `examples/basic/mcp_basic_agent/README.md` - **Simplest Usage of MCP Agent - Hello World!** (documentation): Simplest Usage of MCP Agent - Hello World! Evidence: `examples/basic/mcp_hello_world/README.md` - **LLM Selector example** (documentation): This example shows using MCP's ModelPreferences type to select a model LLM based on speed, cost and intelligence priorities. Evidence: `examples/basic/mcp_model_selector/README.md` - **MCP aggregator example** (documentation): This example shows connecting to multiple MCP servers via the MCPAggregator interface. An MCP aggregator will combine multiple MCP servers into a single interface allowing users to bypass limitations around the number of MCP servers in use. Evidence: `examples/basic/mcp_server_aggregator/README.md` - **MCP Tool Filter Example** (documentation): This example demonstrates a non-invasive approach to filtering MCP tools without modifying any core mcp-agent code. Evidence: `examples/basic/mcp_tool_filter/README.md` - **OAuth Basic MCP Agent example client-only loopback** (documentation): OAuth Basic MCP Agent example client-only loopback Evidence: `examples/basic/oauth_basic_agent/README.md` - **Streaming Demo** (documentation): This example demonstrates the real-time streaming capabilities of mcp-agent, showing how to stream LLM responses as they're generated. Evidence: `examples/basic/streaming_demo/README.md` - **Token Counter Example** (documentation): This example demonstrates the MCP Agent's token counting capabilities with custom monitoring and real-time tracking. Evidence: `examples/basic/token_counter/README.md` - **mcp-agent Cloud Examples** (documentation): The examples in this directory are all ready to be deployed to mcp-agent cloud, where they can then be integrated with any MCP client via SSE. Evidence: `examples/cloud/README.md` - **Cloud Agent Factory Temporal + Custom Workflow Tasks** (documentation): Cloud Agent Factory Temporal + Custom Workflow Tasks Evidence: `examples/cloud/agent_factory/README.md` - **ChatGPT App Example** (documentation): This example demonstrates how to create an MCP Agent application with interactive UI widgets for OpenAI's ChatGPT Apps platform. It shows how to build a coin-flip widget that renders interactive UI components directly in the ChatGPT interface. Evidence: `examples/cloud/chatgpt_apps/basic_app/README.md` - **Setup** (documentation): A basic coin flip component initialized with create-react-app. Evidence: `examples/cloud/chatgpt_apps/basic_app/web/README.md` - **Timer App - ChatGPT App Example** (documentation): ! timer-app https://github.com/user-attachments/assets/7a526501-84c8-4ef5-b784-4b3948790db2 Evidence: `examples/cloud/chatgpt_apps/timer/README.md` - **Setup** (documentation): A basic coin flip component initialized with create-react-app. Evidence: `examples/cloud/chatgpt_apps/timer/web/README.md` - **Hello World Example** (documentation): This example shows a very basic app with a hello world tool call. Evidence: `examples/cloud/hello_world/README.md` - **MCP Server Example** (documentation): This example is an mcp-agent application that showcases how mcp-agent supports the following MCP primitives: Evidence: `examples/cloud/mcp/README.md` - **Observability Example OpenTelemetry + Langfuse** (documentation): Observability Example OpenTelemetry + Langfuse Evidence: `examples/cloud/observability/README.md` - **MCP Agent Server Example Temporal** (documentation): This example demonstrates how to create an MCP Agent Server with durable execution using Temporal https://temporal.io/ . It shows how to build, run, deploy and connect to an MCP server which leverages Temporal workflows for execution. Evidence: `examples/cloud/temporal/README.md` - **CrewAI Tools Integration Example** (documentation): This example demonstrates how to integrate CrewAI tools into MCP Agent workflows. It shows how to use CrewAI's SerperDevTool for web search and FileWriterTool for file operations within an MCP Agent. Evidence: `examples/crewai/README.md` - **Human interactions in Temporal** (documentation): This example demonstrates how to implement human interactions in an MCP running as a Temporal workflow. Human input can be used for approvals or data entry. In this case, we ask a human to provide their name, so we can create a personalised greeting. Evidence: `examples/human_input/temporal/README.md` - **Third-Party Tools Integration Example** (documentation): Third-Party Tools Integration Example Evidence: `examples/langchain/README.md` - **LM Studio Basic Agent Example** (documentation): This example demonstrates using LM Studio with mcp-agent to run local LLMs with full tool calling and structured output support. Evidence: `examples/lm_studio/README.md` - **Elicitation Example** (documentation): This MCP Agent app shows an Agent which has access to a "Booking System" MCP server. This example highlights the elicitation feature, where a tool can pause its execution to ask the user for additional information or confirmation before proceeding. Evidence: `examples/mcp/mcp_elicitation/README.md` - **Deploying the elicitation example to the cloud** (documentation): Deploying the elicitation example to the cloud Evidence: `examples/mcp/mcp_elicitation/cloud/README.md` - **MCP Primitives Example: Using Resources and Prompts** (documentation): MCP Primitives Example: Using Resources and Prompts Evidence: `examples/mcp/mcp_prompts_and_resources/README.md` - **SSE example** (documentation): This example shows how to use an SSE server with mcp-agent. Evidence: `examples/mcp/mcp_sse/README.md` - **MCP Agent example** (documentation): This example shows a basic agent that can connect to an MCP server over SSE with auth headers. Evidence: `examples/mcp/mcp_sse_with_headers/README.md` - **MCP Streamable HTTP example** (documentation): This example shows mcp-agent usage with a Streamable HTTP server using the example server https://github.com/modelcontextprotocol/python-sdk/tree/main/examples/servers/simple-streamablehttp-stateless in the mcp-python repo . Evidence: `examples/mcp/mcp_streamable_http/README.md` - **MCP Websocket example** (documentation): This example shows a basic agent that can connect to an MCP server over websockets Evidence: `examples/mcp/mcp_websockets/README.md` - **MCP Agent Server Examples** (documentation): This directory contains examples of exposing MCP Agent workflows as MCP servers. It demonstrates how to build, launch, and interact with agent-powered MCP servers in different execution environments. Evidence: `examples/mcp_agent_server/README.md` - **MCP Agent Server Example Asyncio** (documentation): This example is an mcp-agent application that is exposed as an MCP server, aka the "MCP Agent Server". Evidence: `examples/mcp_agent_server/asyncio/README.md` - **Context Isolation Demo** (documentation): This example shows how per-request context scoping prevents logs and notifications from bleeding between concurrent MCP clients. Evidence: `examples/mcp_agent_server/context_isolation/README.md` - **MCP Agent Server Example Temporal** (documentation): This example demonstrates how to create an MCP Agent Server with durable execution using Temporal https://temporal.io/ . It shows how to build, run, and connect to an MCP server that uses Temporal as the execution engine. Evidence: `examples/mcp_agent_server/temporal/README.md` - **MCP Azure Agent Example - "Finder" Agent** (documentation): MCP Azure Agent Example - "Finder" Agent Evidence: `examples/model_providers/mcp_basic_azure_agent/README.md` - **MCP Bedrock Agent Example - "Finder" Agent** (documentation): MCP Bedrock Agent Example - "Finder" Agent Evidence: `examples/model_providers/mcp_basic_bedrock_agent/README.md` - **MCP Google Agent Example - "Finder" Agent** (documentation): MCP Google Agent Example - "Finder" Agent Evidence: `examples/model_providers/mcp_basic_google_agent/README.md` - **MCP Ollama Agent example** (documentation): This example shows a "finder" Agent using llama models to access the 'fetch' and 'filesystem' MCP servers. Evidence: `examples/model_providers/mcp_basic_ollama_agent/README.md` - **OAuth Examples** (documentation): Two complementary scenarios demonstrate how OAuth integrates with MCP: Evidence: `examples/oauth/README.md` - **OAuth Interactive Tool Example** (documentation): This example shows the end-to-end OAuth authorization code flow for a simple synchronous MCP tool. The MCP server exposes a github org search tool that calls the GitHub MCP server. When the tool is invoked without a cached token, the server issues an auth/request message and the client opens the browser so you can complete the GitHub sign-in. Evidence: `examples/oauth/interactive_tool/README.md` - **Workflow Pre-Authorize Example** (documentation): This example shows how to seed OAuth credentials for asynchronous workflows. The client calls the workflows-store-credentials tool to cache a token for a specific workflow before the workflow runs. Once the token is saved, the workflow can access the downstream MCP server without further user interaction. Evidence: `examples/oauth/pre_authorize/README.md` - **OAuth protected resource example** (documentation): This example shows how to integrate OAuth2 authentication to protect your MCP. Evidence: `examples/oauth/protected_by_oauth/README.md` - **Temporal Workflow Examples** (documentation): This collection of examples demonstrates how to use Temporal https://temporal.io/ as the execution engine for MCP Agent workflows. Temporal is a microservice orchestration platform that helps developers build and operate reliable applications at scale. These examples showcase various workflow patterns and use cases. Evidence: `examples/temporal/README.md` - **MCP Agent example** (documentation): This example shows tracing integration in a basic "finder" Agent which has access to the 'fetch' and 'filesystem' MCP servers. Evidence: `examples/tracing/agent/README.md` - **Langfuse Trace Exporter Example** (documentation): This example shows how to configure a Langfuse OTLP trace exporter for use in mcp-agent by adding a typed OTLP exporter with the expected endpoint and headers. Following information from https://langfuse.com/integrations/native/opentelemetry Evidence: `examples/tracing/langfuse/README.md` - **MCP Agent example** (documentation): This example shows tracing integration for AugmentedLLMs. Evidence: `examples/tracing/llm/README.md` - **SSE example** (documentation): This example shows distributed tracing between a client and an SSE server. mcp-agent automatically propagates trace context in the client requests to the server; the server should be instrumented with opentelemetry and have MCPInstrumentor auto-instrumentation configured from openinference-instrumentation-mcp . Evidence: `examples/tracing/mcp/README.md` - **Temporal Tracing Example** (documentation): This example demonstrates how to use Temporal https://temporal.io/ as the execution engine for MCP Agent workflows, with OpenTelemetry tracing enabled. Evidence: `examples/tracing/temporal/README.md` - **FastAPI WebSocket Example with MCP Agent** (documentation): FastAPI WebSocket Example with MCP Agent Evidence: `examples/usecases/fastapi_websocket/README.md` - **marimo MCP Agent example** (documentation): This example marimo https://github.com/marimo-team/marimo notebook shows a "finder" Agent which has access to the 'fetch' and 'filesystem' MCP servers. Evidence: `examples/usecases/marimo_mcp_basic_agent/README.md` - **🌐 Browser Console Agent Example** (documentation): A command-line application that lets you interact with websites using natural language through the Model Context Protocol MCP with the use of the Puppeteer MCP server https://github.com/modelcontextprotocol/servers/tree/main/src/puppeteer . Evidence: `examples/usecases/mcp_browser_agent/README.md` - **MCP Financial Analyzer with Google Search** (documentation): MCP Financial Analyzer with Google Search Evidence: `examples/usecases/mcp_financial_analyzer/README.md` - **GitHub PRs to Slack Summary Agent** (documentation): This application creates an MCP Agent that monitors GitHub pull requests and submits prioritized summaries to Slack. The agent uses a LLM to analyze PR information, prioritize issues, and create informative summaries. Evidence: `examples/usecases/mcp_github_to_slack_agent/README.md` - **Instagram Gift Advisor** (documentation): An MCP Agent that analyzes Instagram profiles to generate personalized gift recommendations with real Amazon product links. Evidence: `examples/usecases/mcp_instagram_gift_advisor/README.md` - **MCP Marketing Content Agent** (documentation): This example demonstrates a marketing content creation agent that learns your brand voice and generates platform-optimized content using an evaluation-driven approach with persistent memory for continuous improvement. Evidence: `examples/usecases/mcp_marketing_assistant_agent/README.md` - **LinkedIn Candidate Search & CSV Export Tool** (documentation): LinkedIn Candidate Search & CSV Export Tool Evidence: `examples/usecases/mcp_playwright_agent/README.md` - **MCP Research & Analysis Agent Framework** (documentation): MCP Research & Analysis Agent Framework Evidence: `examples/usecases/mcp_realtor_agent/README.md` - The remaining 20 evidence entries are in `AI_CONTEXT_PACK.json` or `EVIDENCE_INDEX.json`. ## Rules the Host AI Must Follow - **Treat this asset as pre-work context, not a runtime environment.**: The AI Context Pack contains only an evidence-backed understanding of the project, not the project's executable state. Evidence: `docs/README.md`, `README.md`, `examples/basic/agent_factory/README.md` - **When answering the user, distinguish what can be previewed from what can only be verified after install.**: The consumer value of the pre-install experience comes from reducing bad installs and misjudgments, not from pretending to be a real run. Evidence: `docs/README.md`, `README.md`, `examples/basic/agent_factory/README.md` ## Questions the User Should Answer First - Which host AI or local environment do you plan to use it in? - Do you just want to experience the workflow first, or are you ready to actually install? - What matters most to you: install cost, output quality, or conflicts with your existing rules? ## Acceptance Checks - Every capability claim can be traced back to a file path in evidence_refs. - AI_CONTEXT_PACK.md does not package previews as a real run. - The user can understand who it fits, what it can do, how to start, and the risk boundaries within 3 minutes. --- ## Doramagic Context Augmentation The following sections strengthen the repository context for a host AI. Human Manual data is a reading route, and pitfall notes become operating constraints. ## Human Manual Outline Usage rule: this is only a reading route and salience signal, not factual authority. Concrete claims must still return to repo evidence or Claim Graph. Host AI hard rules: - Do not treat page titles, section order, summaries, or importance values as factual project evidence. - When explaining the Human Manual outline, state that it is only a reading route or salience signal. - Capability, installation, compatibility, runtime state, and risk claims must cite repo evidence, source paths, or Claim Graph. - **Introduction and System Architecture**: importance `high` - source_paths: README.md, pyproject.toml, src/mcp_agent/app.py, src/mcp_agent/__init__.py, src/mcp_agent/core/context.py - **Core Components: MCPApp, Agents, and Augmented LLMs**: importance `high` - source_paths: src/mcp_agent/app.py, src/mcp_agent/agents/agent.py, src/mcp_agent/agents/agent_spec.py, src/mcp_agent/workflows/llm/augmented_llm.py, src/mcp_agent/workflows/llm/augmented_llm_openai.py - **Workflow Patterns and MCP Server Integration**: importance `high` - source_paths: src/mcp_agent/workflows/factory.py, src/mcp_agent/workflows/parallel/parallel_llm.py, src/mcp_agent/workflows/router/router_llm.py, src/mcp_agent/workflows/orchestrator/orchestrator.py, src/mcp_agent/workflows/evaluator_optimizer/evaluator_optimizer.py - **Production Features: Temporal, Cloud, OAuth, and Observability**: importance `high` - source_paths: src/mcp_agent/executor/workflow.py, src/mcp_agent/executor/temporal/temporal_context.py, src/mcp_agent/cli/cloud/main.py, src/mcp_agent/cli/cloud/commands/deploy/main.py, src/mcp_agent/oauth/flow.py ## Repo Inspection Evidence - repo_clone_verified: true - repo_inspection_verified: true - repo_commit: `f62d849350816588b1c6294e7914bbe4d8b84072` - inspected_files: `README.md`, `pyproject.toml`, `uv.lock`, `docs/README.md`, `docs/advanced/composition.mdx`, `docs/advanced/monitoring.mdx`, `docs/advanced/temporal.mdx`, `docs/cloud/authentication/deployment-auth.mdx`, `docs/cloud/authentication/external-mcp-auth.mdx`, `docs/cloud/authentication/overview.mdx`, `docs/cloud/deployment-quickstart.mdx`, `docs/cloud/mcp-agent-cloud/deploy-mcp-server.mdx`, `docs/cloud/mcp-agent-cloud/long-running-tools.mdx`, `docs/cloud/mcp-agent-cloud/manage-secrets.mdx`, `docs/cloud/mcp-agent-cloud/overview.mdx`, `docs/cloud/mcp-agent-cloud/use-deployed-server.mdx`, `docs/cloud/observability.mdx`, `docs/cloud/overview.mdx`, `docs/cloud/use-cases/build-chatgpt-apps.mdx`, `docs/cloud/use-cases/deploy-agents.mdx` Host AI hard rules: - Without repo_clone_verified=true, do not claim that the source code has been read. - Without repo_inspection_verified=true, do not write README, docs, or package-file conclusions as facts. - Without quick_start_verified=true, do not claim that the Quick Start path has run successfully. ## Doramagic Pitfall Constraints These rules come from Doramagic discovery, validation, or compilation findings. The host AI must treat them as operating constraints, not background notes. - No project-specific pitfall log has English-native content; rely on the base pack's risk card and evidence index. Do not invent missing facts.