Help us improve
Share bugs, ideas, or general feedback.
From sessionlog
Exports current Claude Code session to standard LLM conversation JSON and human-readable TXT transcript via bash script. Activates on requests like 'export session log' or 'save session'.
npx claudepluginhub florianbuetow/claude-code --plugin sessionlogHow this skill is triggered — by the user, by Claude, or both
Slash command
/sessionlog:exportThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Export the current Claude Code session to standard LLM conversation JSON and a human-readable TXT transcript.
Exports Claude Code session JSONL files to human-readable markdown with role markers and timestamps. Use to save, review, share transcripts, or feed dev-diary pipelines.
Saves the current Claude Code session transcript as clean markdown (conversation only) and raw transcript files for reuse in memos, blog drafts, or articles. Supports starting, refreshing, and finalizing logs per session.
Exports Claude Code session transcripts to GitHub-flavored Markdown with metadata header, collapsible tool results, and thinking blocks. Invoke via /export-session-as-markdown for current or specified path.
Share bugs, ideas, or general feedback.
Export the current Claude Code session to standard LLM conversation JSON and a human-readable TXT transcript.
Run the following to find the session file:
project_dir="$HOME/.claude/projects/$(pwd | sed 's|/|-|g')"
current_session=$(ls -t "$project_dir"/*.jsonl 2>/dev/null | head -1)
session_id=$(basename "$current_session" .jsonl)
echo "Session: $session_id"
echo "Source: $current_session"
Use the output directory specified in the user's prompt. If none was specified, default to docs/sessionlogs/ relative to the current working directory.
"${CLAUDE_PLUGIN_ROOT}/scripts/export-session.sh" "$current_session" "<output-dir>" "$session_id"
The script produces two files:
<output-dir>/claude-<session-id>.json — standard LLM conversation JSON (array of {role, content} objects)<output-dir>/claude-<session-id>.txt — human-readable transcript with [user]/[assistant] prefixesShow the user:
JSON — Array of message objects following the standard LLM conversation format:
[
{"role": "user", "content": "Hello"},
{"role": "assistant", "content": [{"type": "text", "text": "Hi there!"}]}
]
TXT — Human-readable transcript:
Session: <session-id>
[user] Hello
[assistant] Hi there!