Search AI coding agent histories with filtering and context expansion.
npx skills add https://github.com/minhtri2710/dotfiles --skill cass-minhtri2710
Please help me install this Agent Skill. Skill: cass Source: https://github.com/minhtri2710/dotfiles/tree/main/dot_config/opencode/skill/cass Command: npx skills add https://github.com/minhtri2710/dotfiles --skill cass-minhtri2710
Cross-Agent Session Search (CASS) lets you search across all your AI coding agent histories before solving problems from scratch. It helps you reuse past solutions, compare approaches, and avoid reinventing the wheel.
cass_search(query="authentication token refresh", limit=5) cass_view(path="/path/from/search", line=42) cass_expand(path="/path", line=42, context=10)
High-intent search queries and answers about installing and using this skill.
Cross-agent session search lets you query all agent histories to surface prior solutions. Use cass_search(query="your search term", limit=5) to find relevant code, decisions, and approaches from past sessions before implementing from scratch.
Yes. Search your agent histories to answer "what did I try last time" when facing errors. cass_search retrieves past debugging attempts, solutions, and error patterns across all sessions to accelerate troubleshooting.
Use cass_view(path="/path/from/search", line=42) to inspect specific matches, then cass_expand(path="/path", line=42, context=10) to retrieve surrounding code and understand the full decision context.
Search agent histories to identify similar problems solved before, compare approaches different agents took on the same APIs, and extract patterns. This avoids reinventing solutions and surfaces proven techniques from your history.
Yes. Cross-agent search supports filtering by agent and time to narrow results. Combine cass_search with these filters to locate solutions from particular sessions or within specific date ranges.
No dependencies or components are required. Cross-agent search works directly on your existing AI agent histories with no prior setup—query immediately to surface past solutions and decisions.
brucebanner010198-commitsAppend per-agent dispatch and report transcripts to JSONL session logs.
vib795Detect drift between documented learnings and current code, then refresh affected pattern docs.
gil-kapelAnalyze log and code error patterns to identify root causes across distributed systems.
IAmKongHaiSearch and query coding agent session history for past activities and outcomes.
