菜单
Use when optionally installing or using semble_rs for semantic codebase exploration, dependency/impact hints, and compact build/test/CI log digests inside Hermes workflows.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | semble-rs |
| description | Use when optionally installing or using semble_rs for semantic codebase exploration, dependency/impact hints, and compact build/test/CI log digests inside Hermes workflows. |
| version | 0.1.0 |
| author | Hermes Agent |
| license | MIT |
| metadata | {"hermes":{"tags":["code-search","logs","rust-cli","token-efficiency"],"related_skills":["hermes-memory-stack"]}} |
semble_rs is an optional CLI for agent-friendly code exploration and log compression. It complements Hermes search_files and read_file; it does not replace grounded file reads.
cargo install --git https://github.com/johunsang/semble_rs --locked --force
ln -sf "$HOME/.cargo/bin/semble_rs" "$HOME/.local/bin/semble_rs"
semble_rs plan "where is tool dispatch implemented" /path/to/repo -k 5
semble_rs search "context compression routing" /path/to/repo --outline -k 8
pytest 2>&1 | semble_rs digest
cargo test 2>&1 | semble_rs digest
cargo is absent, skip this integration.semble_rs --help works.printf 'error: test' | semble_rs digest returns compact output.Use when installing, configuring, or explaining the Hermes Memory Stack: a layered optional memory/input bootstrap for Hermes Agent with skills, Axiom Perception MCP, llmwiki, Hindsight routing, optional semble_rs, and optional NotebookLM.
Use when installing or using Axiom Perception MCP as a procedural execution-pattern cache for Hermes Agent UI, browser, and desktop workflows.
Use when configuring or using llmwiki as a shared agent knowledge base, including API-free markdown mode, OpenAI-compatible extraction, local/OAuth bridge extraction, and optional embedding endpoints.
Use when optionally installing or using NotebookLM CLI as an external Google-account source-grounded research workspace for PDFs, URLs, YouTube, manuals, and reference packs; not core Hermes memory.