glance

xtftbwvfp/glance

★ 0 stars Rust 🤖 AI/LLM Updated today

MCP server that delegates research / md / obsidian work to a cheap sub-agent (GLM, DeepSeek, OpenAI-compat). Saves codex/claude/cursor tokens on the heavy 'read N files' loop.

View on GitHub →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "glance": {
      "command": "cargo",
      "args": [
        "run",
        "--",
        "glance"
      ]
    }
  }
}

README Excerpt

> An MCP server that delegates **research / markdown / obsidian** work to a > cheap sub-agent (GLM, DeepSeek, OpenAI, …) so codex / claude / cursor save > tokens on the heavy "read N files and figure out what's there" loop. When you ask codex "find the login validation logic", codex itself reads a dozen files into its 200K context window. By turn 30 the context is full and

Tools (17)

explainfix_lintimage_describemd_outlinemd_readmd_writeobsidian_backlinksobsidian_readobsidian_searchobsidian_writerepo_exploreresearchsearchweb_fetchweb_searchwrite_docswrite_tests

glance

Quick Install

README Excerpt

Tools (17)

Related AI/LLM Servers