$ loading_
Orchestrate AI agent workflows with dependencies, parallel execution, and failure policies.
This MCP tool has limited documentation, declares no required secrets or remote endpoints, and is open source and auditable, with no clear high-risk red flags observed. It is known to execute code locally, and its weak adoption and unknown maintenance status warrant cautious use.
The materials explicitly state that no API keys or environment variables are required, and there is no stated design for token collection, storage, or credential forwarding; based on available facts, credential exposure appears low.
No remote endpoints or external hosts are declared, and the materials do not describe sending user data to third-party services; based on current disclosure, no explicit network egress path is evident.
The system checks explicitly indicate this tool has code-execution capability; combined with its stated role in 'campaign orchestration / dependency DAGs / parallel fan-out,' it is reasonable to infer that it orchestrates and triggers local tasks or processes. This is a typical high-privilege MCP capability and should be used in a controlled environment.
Documentation is absent, and the scope of readable/writable files, directories, or other local resources is not clearly specified; given its orchestration role and code-execution capability, it would typically interact with task inputs/outputs and local state data. There is no concrete evidence of overbroad access, but the permission boundary is unclear.
The project is open source and GPL-3.0 licensed, making the code auditable in principle, which is a meaningful risk-reducing factor; however, it comes from a third-party registry, has 0 stars, and has unknown maintenance status, so public trust signals are weak and supply-chain confidence should remain conservative.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "sortie-mcp" yet — see the docs or source repo.
Create an orchestration for a product release: first generate release notes, then run documentation updates, test checks, and marketing asset preparation in parallel; if testing fails, stop the remaining release steps and attach notes to each node.
A release workflow plan with dependencies, parallel branches, failure handling rules, and node notes.
Build a DAG for a marketing campaign: first define the target audience, then generate email copy, ad creatives, and a landing page draft in parallel; if any content review fails, automatically route back to a revision node and record the reason.
A campaign task DAG with parallel fan-out, review-failure fallback policies, and embedded notes.
Plan a research agent workflow: first gather problem context, then search sources, extract key points, and draft initial conclusions in parallel; if search results are insufficient, automatically add follow-up search steps and preserve process notes.
An orchestrated research pipeline showing dependencies, parallel steps, exception handling, and attached notes.
Control agent workflows with stateful primitives and persisted execution facts.
Run dependent or parallel agent tasks and return structured results in one call.
Orchestrate multiple AI agents in real time and monitor tasks and artifacts.
Automate task breakdown, dependency tracking, and smart project recommendations.
Orchestrate local multi-agent workflows with gated lifecycle, handoffs, and host continuation.
Enable structured AI-human messaging and Q&A orchestration through Telegram.