From watercooler
Finds discussions and entries related to a topic or entry ID using semantic search. Useful for discovering connected context across threads.
How this skill is triggered — by the user, by Claude, or both
Slash command
/watercooler:find-relatedThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Find content related to: $ARGUMENTS
Find content related to: $ARGUMENTS
Determine argument type:
01HQXYZ123ABC456DEF789GHJ) — exactly 26 characters matching ^[0-9A-HJKMNP-TV-Z]{26}$ (Crockford base32)For Entry ID — use seeded similarity search:
ToolSearch: select:mcp__watercooler__watercooler_search
Then call:
mcp__watercooler__watercooler_search(seed_entry_id="<ulid>")
For Description — use semantic search:
ToolSearch: select:mcp__watercooler__watercooler_search
Then call:
mcp__watercooler__watercooler_search(query="<description>", mode="entries", semantic=true, code_path="<repo root>")
Present results:
Suggest deeper exploration:
/find-related 01HQXYZ123ABC456DEF789GHI - Find entries similar to this one/find-related git sync conflict resolution - Find related discussions/find-related branch parity implementation - Discover connected contextnpx claudepluginhub mostlyharmless-ai/watercooler --plugin watercoolerCompiles deep context on a topic by combining semantic search with relationship traversal across 4 retrieval phases. Use for deep research on topics spanning multiple entries and relationships.
Searches threads with filters including role, type, date range, thread topic, and status. Supports AND/OR token matching for multi-keyword queries.
Search previous Claude Code conversations for facts, patterns, decisions, and context using semantic or text search.