Yasin-MCP
Provides read-only access to GitHub repository state information, allowing AI agents to query repository data such as project status and metadata as part of the Yasin ecosystem.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Yasin-MCPWhat's in the Yasin project registry?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Yasin-MCP
Standalone AI/Agent-facing MCP (Model Context Protocol) access and integration layer for the Yasin ecosystem.
Status
Yasin-MCP roadmap Stages 1–15 are complete on the controlled-integration path. Master #96 is the roadmap closure gate. Stage 15 is the final planned roadmap stage; no planned Stage 16 remains.
The repository has reached a stable v1.0.0 release for the currently verified scope. The release is suitable for controlled integration. This classification does not claim a public package publication or unrestricted production deployment.
Roadmap completion evidence
Stage 11 / #97: merged.
Stage 12 / #98 → PR #103: merged.
Stage 13 / #99 → PR #104: merged; live ecosystem-compatible MCP client path verified over authenticated Streamable HTTP, including identity propagation, governance, approval, execution, audit correlation, and fail-closed malformed context handling.
Stage 14 / #100 → PR #105 and lifecycle-resilience PR #106: merged; bounded concurrency, deterministic lifecycle/isolation coverage, stress behavior, cleanup, slot recovery, and post-stress usability are covered.
Stage 15 / #101: final release, reproducibility, compatibility, documentation, transport, security, and roadmap-closure verification.
Current verification
The latest controlled Termux verification completed successfully:
402 tests passed
2 tests skipped
Native Termux environment verified with Python 3.14.6
cryptography 50.0.1imported successfully in the verified environmentOperations subprocess transport verified using a Termux-compatible Python gateway invocation
See docs/RELEASE_READINESS.md and docs/compatibility/YASIN_MCP_TERMUX_COMPATIBILITY.md for the detailed assessment and Termux evidence.
Related MCP server: agenticscope
Architecture boundary
Yasin-MCP does not replace YASIN-DOCS, Yasin-Core, Yasin-Agent, Yasin-AI, YasinHub, YasinCLI, or Yasin-Operations. It is an access/integration layer.
MCP capabilities are read-only at the product surface unless an explicitly governed reference capability is present for testing governance behavior.
No generic shell passthrough or arbitrary command execution is exposed.
Integrations use public APIs, SDKs, contracts, or explicit adapters; no private cross-repository imports are required.
Tool execution crosses the centralized
GovernanceGatefor authentication, approval, policy, audit, and bounded-concurrency enforcement.External/untrusted content is represented using explicit evidence/trust boundaries and is not treated as instructions.
Tool surface
Always available:
yasin_docs_*— documentation accessyasin_github_*— read-only GitHub ecosystem accessyasin_registry_*— project/dependency registry accessyasin_gov_*— governance reference capabilities used to exercise policy/approval boundaries
Conditionally available:
yasin_operations_*— registered only when theyasin-operationsexecutable is available onPATH.
The capability surface is versioned independently through CAPABILITY_SURFACE_VERSION.
Evidence model
Responses that report ecosystem state use one of:
CONFIRMED— directly observed from a live authoritative sourceTARGET— documented intent/architecture, not verified livePROPOSED— suggestion or plan, not implementedUNRESOLVED— could not be determined; never present this as fact
Security and governance
Yasin-MCP treats documentation, GitHub content, registry content, and Operations output as external/untrusted data. Structural trust/evidence envelopes preserve the distinction between retrieved data and instructions.
The governance path is centralized and fail-closed. Authentication is established at the boundary; approval is explicit for mutation-risk reference capabilities; policy decisions are audited; structured errors are used at public boundaries; and configured concurrency is bounded.
Transport
stdio
The standard MCP stdio transport is supported and covered by the repository's live client/CLI validation surface.
Streamable HTTP
Remote transport is implemented through an ASGI application and supports bearer authentication. Stage 13 provides live local verification using the official MCP Python client, with a real ClientSession, Streamable HTTP connection, authentication, context propagation, governance, approval, execution, and audit correlation.
Remote deployments require TLS unless remote_allow_insecure_http is explicitly enabled for local testing.
Packaging and supported runtimes
pyproject.toml declares:
package version:
0.1.0(the package metadata version remains independent from the Git release tag)requires-python = ">=3.10"mcp>=2,<3PyYAML>=6,<7
CI validates Python 3.10, 3.11, and 3.12 with Ruff, Mypy, Bandit, and pytest.
Termux
Native Termux compatibility is verified for the current repository state. The verified environment is Android/Termux on aarch64 with Python 3.14.6. The repository includes a dedicated scripts/termux-test.sh compatibility check and the Operations adapter supports the Termux Python-gateway invocation model.
Termux verification is an environment-specific compatibility result; it does not change the declared general CPython requirement above or imply that every Android/Termux package combination is supported.
Development
pip install -e ".[dev]"
pytest
ruff check .
ruff format --check .
mypy src
bandit -q -r srcFor Termux verification:
source .venv/bin/activate
bash scripts/termux-test.shDocumentation index
Doc | Purpose |
Architecture and evidence map | |
Install, run, diagnose, and failure modes | |
Final roadmap/release-readiness assessment | |
Historical Stage 5 assessment | |
Governance model | |
Stdio client configuration and smoke checklist | |
Capability surface/version semantics | |
Live runtime evidence | |
Optional Operations gateway | |
Registry consumer contract | |
Retry/reliability policy | |
Correlation and redaction | |
Notable changes |
Final roadmap rule
Master #96 defines the current five-issue completion roadmap. Stage 15 is its final planned stage. Future maintenance or genuinely new product features may be added later, but omitted roadmap work must not be hidden under a new Stage 16.
Native Termux compatibility is explicitly documented as verified for the current repository state based on the controlled test evidence above.
Maintenance
Related MCP Connectors
Read-only MCP tools for AI agent discovery, structured resources, and NIULAI information.
The Cortex MCP server provides read-only access to real-time engineering context from the Cortex developer portal, allowing AI coding assistants to answer natural language questions about your organization's catalog (microservices, libraries, domains, teams, infrastructure), scorecards (engineering standards and best practices), initiatives (goals and deadlines), and Engineering Intelligence metrics. It includes tools for querying documentation, tracking personal entities, and accessing AI-assisted insights across the entire Cortex ecosystem.
Agent-native MCP server over the public saagarpatel.dev corpus. Read-only, stateless.
Read-only MCP server exposing a user ORANO library to their own AI agent.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceA read-only MCP server for AI coding agents to inspect repositories, audit code quality, route engineering skills, and plan safe issue/PR workflows.1MIT
- AlicenseNot gradedqualityAmaintenanceA read-only MCP server that provides AI agents with live, structured workspace awareness, including project listing, git status, and budgeted context packing, minimizing token usage.82MIT
- AlicenseAqualityCmaintenanceRead-only MCP server providing AI access to verifiable web, GitHub, and local sources, plus a managed fantasy entity catalog, with strong security and provenance tracking.101MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to query the Canton Network ecosystem through a unified MCP endpoint, providing live data from forums, GitHub, docs, CIPs, and more with read-only access.Apache 2.0