Index and search coding agent session histories with JSON output.
npx skills add https://github.com/mistakenot/auto-stack --skill cass-mistakenot
Please help me install this Agent Skill. Skill: cass Source: https://github.com/mistakenot/auto-stack/tree/main/.agents/skills/cass Command: npx skills add https://github.com/mistakenot/auto-stack --skill cass-mistakenot
Aggregates and indexes scattered coding agent session logs so developers and AI agents can quickly find prior interactions, debug recurring issues, and transfer knowledge across different agent platforms.
Search your coding agent history for "authentication error" and return up to five machine-readable JSON results including path, line, agent, title, snippet, and score.
High-intent search queries and answers about installing and using this skill.
You can search coding agent session histories by indexing local logs from Codex, Claude Code, Gemini, Cursor, and Aider. This provides unified lexical and semantic search, returning snippets and scores for rapid retrieval across all platforms.
Transferring context to a new agent requires filtering past interactions by agent, workspace, or time. You can locate exact solved bug snippets and export them as machine-readable JSON for automated downstream handoff.
Yes, you can use optional local semantic search to find specific code snippets from past AI agent interactions. It indexes session stores to retrieve relevant context, applying token-budget controls for LLM integration.
Yes, robot mode outputs machine-readable JSON for coding agent history searches. It includes path, line, agent, title, snippet, and score, supporting robust filtering and pagination for automated agent consumption.
You index and deduplicate session logs from multiple coding agents by running incremental or full reindexing. Watch mode provides live updates, ensuring your filtered corpus remains current and clean for cross-agent knowledge transfer.
Limitations of searching local coding agent session stores include managing token budgets for LLM integration and handling incremental reindexing overhead. Robust diagnostics help mitigate issues when deduplicating across diverse agent formats.
lucking7Search and filter coding agent session history across Claude Code, Codex, and Cursor.
ataskiteInteract with the AgentScope Demo AI agent server via command-line interface.
trungtnmIndex and search local coding agent histories across multiple agents.
nahisahoSearch the RCSB PDB by text query and retrieve structured metadata with ligand information.
