- name
SKILL.md
技能檔案保留在發布者的儲存庫中。
在終端機中貼上複製的指令。它會將這個 SKILL.md 安裝到對應的代理中。
npx skills add axoviq-ai/synthadoc --skill session -a cursor -y在終端機中貼上複製的指令。它會將這個 SKILL.md 安裝到對應的代理中。
npx skills add axoviq-ai/synthadoc --skill session -a claude-code -y在終端機中貼上複製的指令。它會將這個 SKILL.md 安裝到對應的代理中。
npx skills add axoviq-ai/synthadoc --skill session -a codex -y在終端機中貼上複製的指令。它會將這個 SKILL.md 安裝到對應的代理中。
npx skills add axoviq-ai/synthadoc --skill session -a grok -y在 Grok Bot 聊天中傳送複製的訊息。Bot 會從 GitHub 取得這個 SKILL.md,並加入你所有 Bot 共用的技能中。
Install this skill: https://github.com/axoviq-ai/synthadoc/tree/HEAD/synthadoc/skills/session來自發布者
Extracts human-readable conversation turns from AI coding session history files
(.jsonl). Supports two formats:
~/.claude/projects/<hash>/<session-id>.jsonl){"role": ..., "content": ...} per-line format
used by OpenAI Codex and Cursor IDE sessionsFormat is detected automatically from the first parseable line.
Only substantive conversation turns are kept:
The extracted text is then passed through Synthadoc's standard pre-LLM source sanitizer (zero-width characters, bidi overrides, HTML comments, hidden CSS spans, base64 blobs, instruction-override phrases), exactly like PDF, DOCX, URL, and every other source type.
Each turn is labelled [USER] or [ASSISTANT] and separated by ---:
[USER]
How do I implement a sliding window algorithm?
---
[ASSISTANT]
A sliding window algorithm maintains a contiguous subarray (the "window") …
suggested_slugThe skill returns a suggested_slug in metadata derived from the session file's
modification time and the first substantive user message:
session-2026-07-15-how-do-i-implement-a-sliding
Sessions longer than 30 substantive turns are split into 30-turn chunks.
Each chunk is labelled with a ## Part N of M header so the downstream LLM
can process sections independently. The metadata dict includes chunk_total
when chunking occurs; single-chunk sessions (≤ 30 turns) are unchanged.
ExtractedContent..jsonl"claude session", "codex session", "cursor session",
"ai session", "session history"import asyncio
from synthadoc.skills.session.scripts.main import SessionSkill
skill = SessionSkill()
async def main():
result = await skill.extract("/path/to/session.jsonl")
print(result.text) # [USER]\n...\n\n---\n\n[ASSISTANT]\n...
print(result.metadata) # {"format": "claude_code", "turn_count": 42, "suggested_slug": "..."}
asyncio.run(main())
Extract conversation turns from AI session history files (.jsonl)
---name: sessiondescription: Extract conversation turns from AI session history files (.jsonl)---# synthadoc/skills/session/SKILL.md常見問題
從發布者下載 synthadoc/skills/session/SKILL.md,或複製路徑後開啟儲存庫。PlugMCP 只收錄這項技能,不代管檔案。
發布者將它存放在 https://github.com/axoviq-ai/synthadoc。
Claude、Cursor、Codex 和 Grok Build 可以讀取你從發布者儲存庫加入的 SKILL.md。在聊天中傳送技能的 GitHub 資料夾連結,Grok Bot 就會安裝它。