| 1234567891011121314151617181920212223 |
- name: tameru
- title: Tameru
- repo: https://github.com/0xWhiteMage/hermes-tameru-plugin
- sha: 9f80514ec93cf1973601e39811980b823cb2a766
- subdir: hermes_tameru_plugin
- version: "1.6.0"
- description: "Big tool outputs, kept small, nothing lost: old tool results shrink to the lines that matter and every distinct error, and the tameru_expand tool brings back the exact original. Deterministic and local, with no LLM calls. Disclosure — Subclasses Hermes's built-in ContextCompressor and overrides its private prune/demote methods (checks their shape at startup and falls back to stock compaction when they change). Unless you set compression.proactive_prune_tokens yourself, turns on proactive tool-result pruning with a trigger scaled to the model's context window, and after each prune shortens Hermes's re-prune runway. Adds a model-callable tameru_expand tool that returns pruned tool output from memory or from this session's archive (web, browser and MCP output stays wrapped as untrusted data); complex regex greps run in a short-lived local Python child process. No network calls; the render log is off by default."
- maintainer: 0xWhiteMage
- tier: community
- category: general
- docs_url: https://github.com/0xWhiteMage/hermes-tameru-plugin
- image: https://raw.githubusercontent.com/0xWhiteMage/hermes-tameru-plugin/9f80514ec93cf1973601e39811980b823cb2a766/docs/assets/hero.png
- platforms: [linux]
- known_issues:
- - Full archive recovery requires plugins.isolation to be in_process; host mode limits recovery.
- - Overrides private ContextCompressor signatures and falls back when the host contract changes.
- capabilities:
- provides_tools:
- - tameru_expand
- provides_hooks: []
- provides_middleware: []
- requires_env: []
- requires_hermes: ">=0.21.7"
|