Skip to main content
Glama
427,864 tools. Updated 2026-08-09 22:24

"Sage" matching MCP tools:

  • Create detailed implementation plans by leveraging AI models to debate, critique, and refine strategies. Input a task and absolute paths to generate a consensus plan, optionally saving results and full transcripts for future reference.
  • Send code and receive expert review suggestions with SEARCH/REPLACE edits. Specify absolute paths for files or directories to include in the review context. Ideal for detailed code improvements.
  • Send a prompt with absolute file paths to a sage-like model for its opinion or code review. Embed relevant file contents for context to handle large codebases effectively.
  • Analyze a failed SAGE command to identify error type, root cause, and affected files. Obtain actionable suggestions without running any new commands.
    MIT
  • After a SAGE command fails, get safe fix suggestions with explanations and confidence scores without executing any commands.
    MIT
  • Execute one local shell command with a clear purpose label (read, search, test, build, deploy, audit, unknown) to get exit code, compact output, and agent metadata for tracked execution.
    MIT

Matching MCP Servers

  • A
    license
    A
    quality
    A
    maintenance
    Local-first MCP server and CLI wrapper for AI coding agents. SAGE routes shell commands through a tracked local runner, stores command history on the user’s machine, and returns compressed terminal output to reduce noisy context.
    26
    16
    10
    MIT
  • A
    license
    -
    quality
    D
    maintenance
    Enables Claude Desktop and MCP-compatible agents to formulate, solve, and certify mathematical optimization problems using production-grade open-source solvers, providing mathematically grounded decisions.
    MIT

Matching MCP Connectors

  • Run local commands with automatic failure recovery. Executes a command, diagnoses errors, and retries based on autonomy level (suggest, ask, or auto). Ideal for development tasks like tests and builds.
    MIT
  • Retrieve recent local command runs to find a run_id for viewing raw output, explaining errors, or suggesting fixes. Displays compact metadata like exit code, duration, and timestamp.
    MIT
  • Run a local project workflow like test or build, getting structured results with per-step status and duration.
    MIT
  • Find files by pattern, returning newest first, ignoring junk directories, and capping output for concise results.
    MIT
  • Read a file with size-aware compression: returns exact content for small files, or an outline with head and line references for large files to conserve tokens.
    MIT
  • Search files with compressed, grouped results, preserving exact paths and line numbers. Useful for large outputs to reduce noise.
    MIT
  • Retrieve the exact original output of a previous run by ID, restoring full details that compression normally hides.
    MIT
  • Generate a compact directory tree overview with configurable depth and file limit, reducing noise from recursive listings.
    MIT
  • Update Notion page properties, including titles, icons, covers, and status (archived/in trash), using a structured input format. Ideal for programmatically modifying Notion pages via API integration.
    MIT
  • Create and customize pages in Notion with structured content, including titles, blocks, icons, and cover images, using natural language commands to streamline workspace organization.
    MIT