Skip to main content
Glama
Continuous Context Session Compactor

Continuous Context Session Compactor

Official

Continuous Context Session Compactor Skill for Autonomous Agents

MCP Compatible Python 3.9+ License: MIT Zero External Dependencies

Enables long-horizon autonomous agents to run persistently without context window exhaustion. Automatically calculates token consumption and creates concise sliding summaries for elder turns.

Capabilities

  • Token Budget Guarding: Enforces strict configurable budget limits (e.g. 8k, 32k, 128k).

  • Sliding History Distillation: Preserves critical system instructions and recent conversation while compressing historical turns.

  • Zero Overhead: In-memory token estimation and distillation with zero external LLM dependencies.

Related MCP server: cycles-mcp-server

MCP Server Integration

python mcp_server.py

Available MCP Tools

  • compact_messages: Compact conversation array when exceeding token limits.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      B
      maintenance
      Runtime budget authority for autonomous agents - a set of tools to check, reserve, spend, and release budget before and after every costly, risky operation. The agent asks "can I afford this?" before acting, and reports what it actually used afterward.
      9
      57 npm
      Apache 2.0
    • A
      license
      Not graded
      quality
      D
      maintenance
      Virtual Infinite Context for Agents and LLMs - maintains a continuous rolling context window, surfacing relevant memories while respecting token budgets.
      6
      MIT
    • A
      license
      Not graded
      quality
      B
      maintenance
      Provides deterministic context compaction and safety gating for AI agents, preserving original messages verbatim while dropping irrelevant ones and flagging destructive commands.
      2
      MIT