From lore
Distill a Lore thread into handoff context for continuing work. Use when the user asks to continue work from a Lore thread, fork from a thread, pick up from a thread, or use a Lore thread as context for a new task.
How this skill is triggered — by the user, by Claude, or both
Slash command
/lore:fork <lore-thread-url-or-id> <forker-intent><lore-thread-url-or-id> <forker-intent>The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Use the bundled `lore-local` MCP server to generate an intent-conditioned handoff summary for a Lore thread.
Use the bundled lore-local MCP server to generate an intent-conditioned handoff summary for a Lore thread.
Use this skill when the user asks to:
Do not use this for merely reading or listing Lore threads. Use the read skill for read-only thread inspection.
This skill needs two inputs:
th_... and https://lore.tanagram.ai/thread/th_....If either input is missing, ask one concise clarification question before calling the tool.
fork_thread({ thread_id, forker_intent }) on lore-local.source_distilled text as the source-thread context for the current task.The fork result is a JSON object with:
source_thread_id — the source Lore thread id.forker_intent — the intent used for the distillation.source_distilled — the handoff context for the new agent.Do not show raw JSON to the user unless they explicitly ask for it. If the user only asked to prepare the fork context, return source_distilled verbatim. If they asked to continue the work, briefly acknowledge that the fork context was loaded and proceed with the task using that context.
lore_login on lore-local, then retry. If lore_login returns browser_open_failed, tell the user to visit the verification URL, then call lore_login_resume({ device_code }) and retry.source_distilled:
Surface the error and do not invent fork context.Speak about "threads" and "handoff context". Never say "transcript", "JSONL", or "MCP" to the user.
npx claudepluginhub tanagram/lore-plugin --plugin loreCreates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.