ranxianglei/billion-context
Native Hermes hook plugin registering compress/decompress/acp_status tools that route model traffic through billion-context's self-spawned context-compression proxy.
Awesome Hermes Plugins › Context Engines
DECRUX9812/typesafe-skill-router
TypeSafe Skill Router addresses a scaling problem in Hermes Agent: with hundreds of installed skills, each shown to the model as only a one-line description, the model can miss the skill that would actually solve the request. Before the main model call, the plugin sends the current user request together with the names and one-line descriptions of the installed skills, in configurable chunks, to the third-party TypeSafe API (model jev-latest). When a skill clears two independent confidence thresholds - the mean judgment across three internal comparisons and the winning skill's own fit score - the plugin appends one line naming that skill to the user message as a soft hint; the agent is told explicitly that it may ignore the hint if it does not actually fit. The plugin is disabled by default and every failure path - a missing API key, a missing skill roster, a network timeout, or a transport error - is logged once and treated as no suggestion, so a routing failure can never break a Hermes turn. Requests are only routed once the request text is under a length ceiling, and slash commands and multimodal messages are excluded entirely. Its numeric thresholds (gate, fit score, fit margin, shortlist size, chunk size) live in code and are meant to be re-tuned by replaying cached requests, and the README notes that a fit score tuned on English requests scores measurably lower on equivalent non-English requests, which can flip a borderline case from a suggestion to silence. Requirements are Hermes Agent 0.21+, Python 3.10+ with no extra dependencies, and a TypeSafe API key. The only network calls made are to TypeSafe's API, and the plugin documents exactly what request data is sent to that third party.
https://github.com/DECRUX9812/typesafe-skill-router
Also in Context Engines
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.
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...
electricsheephq/lcm-x
Lossless Context Memory eXtension (hermes-lcm-x): a SQLite-backed, DAG-based context engine for Hermes Agent that persists raw messages and compacts them into recoverable summary hierarchies instead of discarding detail.
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.
Elevatormusic/hermes-warm-compaction
Context-engine plugin for Hermes Agent that speeds up compaction by writing the handoff summary against the cached prompt prefix instead of the whole transcript.
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.