context-mode
by mksglu
Keeps raw tool output out of coding agents' context by running code and commands in a sandbox, indexing results for search, and restoring session state after compaction, via MCP.
Skills
Sandboxed Execution
Runs code in 12 languages or processes files in an isolated subprocess with ctx_execute, letting only stdout enter the agent's context.
Indexed Knowledge Base
Chunks markdown or fetched URLs into a SQLite FTS5 index with ctx_index and ctx_fetch_and_index, then returns snippets via ctx_search.
Session Continuity
Tracks file edits, git operations, tasks, errors, and decisions in SQLite, and rebuilds working state after compaction or resume.
Batch Execution
Runs several commands and search queries in one ctx_batch_execute call, with optional concurrency for I/O-bound batches.
Related Agents
Codebase Memory MCP
Indexes repositories into a persistent tree-sitter knowledge graph of functions, classes, call chains, and HTTP routes,…
Graphify
Maps a codebase, docs, PDFs, and media into a queryable knowledge graph so AI coding assistants trace and cite paths in…
LightRAG
Builds a graph-based retrieval index over documents so agents can answer questions spanning many sources, served throug…
AgentOps for Apify Builders Bundle
Apify actor bundle for agent builders: normalize run traces for QA, control costs, and guard tool calls with a firewall…