stephenschoettler/hermes-lcm
hermes-lcm is a lossless context management plugin for Hermes Agent, providing a DAG-based context engine that ensures no messages are lost through persistent storage and intelligent summarization.
Awesome Hermes Plugins › Context Engines
eas4ai/hermes-context-manager
Hermes Context Manager (HMC) is a context optimization plugin for the Hermes Agent gateway. It layers on top of the built-in ContextCompressor rather than replacing it: both run, with HMC acting as a per-tool-call, per-turn silent compressor. The compression pipeline has two tiers. Single-message strategies (pattern short-circuits, code-aware filtering, head and tail truncation) fire as each tool output arrives, so agent loops between user turns stay compressed continuously. Full-list strategies (deduplication, error purging) fire on every user turn and at session end. Code-aware compression strips function, class and struct bodies from outputs containing source, preserving signatures, imports and docstrings; it supports Python with indentation awareness plus Rust, Go, JavaScript and TypeScript, uses string-aware brace counting so braces inside string literals do not confuse it, and bails out on JSX for React safety. Pattern short-circuits collapse recognizable success shapes such as JSON success payloads, pytest and cargo results and git output into one-liners. Deduplication folds tool calls with identical arguments or near-identical normalized content into a single result, with write_file and patch always protected. When context crosses a configurable threshold, background compression asks an auxiliary model, never the main one, to summarize completed work ranges and remove them from the conversation; the summaries land in a searchable per-session index. An opt-in embedded HTTP and SSE server built from the standard library only and bound to localhost streams turn, tool-call and session-end events to a single-page dashboard showing current-session metrics, lifetime totals, a per-strategy breakdown, a recent-sessions table and a live event log. A SQLite store keeps cumulative token savings per session, day, month and project across restarts. Control comes through an hmc_control tool and an /hmc slash command exposing status, context, stats, index, analytics, dashboard, sweep and manual mode actions. It installs through hermes plugins install and needs only Python 3.10 or newer and PyYAML.
https://github.com/eas4ai/hermes-context-manager
Also in Context Engines
claudioemmanuel/squeez
squeez is an end-to-end token optimizer that provides hook-based token compression for Hermes Agent and several other AI CLI hosts, increasing context window efficiency.
EfficientContext/ContextPilot
ContextPilot is a context engine that accelerates long-context LLM inference by optimizing context reuse, offering significant speedups and token savings for RAG, memory chat, and agentic AI worklo...
byenzyme/enzyme-skill
A Hermes plugin that compiles a Markdown vault into a local concept graph and injects trending entities and semantic search results into the session through three lifecycle hooks and five tools.
beniedev/moonbite
Experimental Hermes Agent plugin adding cross-session memory, short-lived working state, and bounded, verified autonomous action on top of the host agent.
kenyonxu/hermes-persona
hermes-persona is a configurable Hermes Agent plugin that dynamically injects persona context into LLM calls using various hooks and rules.
Sahil-SS9/hermes-memlock
Hermes plugin that detects context compaction, audits which pinned standing instructions survived outside the summary region, and rehydrates the lost ones as a reminder block.
PlayForm/Aphrodite-Hermes
Compresses Hermes tool output before it reaches the model into compact retrievable previews, via a thin Python loader calling a Rust classifier and store, cutting per-call token cost.