mcp-context-loop
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
mcp-context-loop
MCP stdio server for CONTEXT LOOP navigation: goto and loop_start_marker.
These tools do not implement context-window mutation themselves. They call
methods angela exposes on a unix-socket host (ANGELA_HOST_SOCK):
HTTP | angela method |
|
|
|
|
Angela starts that host for every harness and passes the socket path into this
server's env when the context-loop MCP preset is connected.
Not enabled by default
New angela agents only get file-io / term-local (or whatever the agent
coffee lists). This server is opt-in:
Add the preset to the agent:
mcp: [
'file-io'
'term-local'
'context-loop'
]Allowlist the tools (they are not on
DEFAULT_LUCY_ALLOWLIST):
goto
loop_start_marker(Registered without a server prefix so the model still sees goto /
loop_start_marker, matching the playbook text.)
Related MCP server: machine-mcp
Tools
Tool | Purpose |
| Bookmark the previous context item; return the per-iteration playbook |
| Hide context after that bookmark (keep the goto nav suffix); return the next-iteration brief |
Env
ANGELA_HOST_SOCK— unix socket path (set by angela). Required for tool calls.
Run (usually via angela)
ANGELA_HOST_SOCK=/tmp/angela-host-….sock bun /workspace/mcp-context-loop/server.mjsTool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
Stop re-explaining yourself to Agents. Give it the right context, right when needed.
Connect AI agents to Replynodes over the Model Context Protocol.
Hosted MCP runtime where your agent publishes its own tools by tool call and keeps working.
Persistent memory, hybrid search and a goal graph for AI agents, over stdio or remote HTTP.
Related MCP Servers
AlicenseNot gradedqualityDmaintenanceEnables local AI agents to communicate directly over Unix sockets, allowing multi-agent collaboration across different models and harnesses without any broker or network dependency.3 npm3Apache 2.0- AlicenseAqualityDmaintenanceEnables an AI agent to run local tools on the user's own machine via stdio, including command execution, workspace file read/write, and system status checks.52MIT
- AlicenseNot gradedqualityBmaintenanceEnables compatible agents to call registered local tools and read resources through one stdio server, including harness preparation, knowledge search and document management, memory retention/recall, correction proposals, advisory decisions, tool/skill discovery, and log/source inspection.2Apache 2.0
- AlicenseBqualityCmaintenanceEnables authenticated clients to discover and call tools for scoped file listing/searching, request client-side LLM sampling, and receive real-time progress/log notifications over stdio or secure Streamable HTTP, with tenant/user workspace isolation and a loopback browser dashboard.7MIT