From workbench-core
Log the current session segment right now — dump the raw log and write the narrative summary + any decision promotions inline. Use this when you want to snapshot mid-conversation, or when you want a richer summary than the auto-generated one.
npx claudepluginhub mike-bronner/claude-workbench --plugin workbench-coreThis skill uses the workspace's default tool permissions.
This is an execution-aware skill — check `skills/log-now.learnings.md` in the vault before proceeding. If it exists, apply accumulated learnings.
Expert guidance for Next.js Cache Components and Partial Prerendering (PPR). **PROACTIVE ACTIVATION**: Use this skill automatically when working in Next.js projects that have `cacheComponents: true` in their next.config.ts/next.config.js. When this config is detected, proactively apply Cache Components patterns and best practices to all React Server Component implementations. **DETECTION**: At the start of a session in a Next.js project, check for `cacheComponents: true` in next.config. If enabled, this skill's patterns should guide all component authoring, data fetching, and caching decisions. **USE CASES**: Implementing 'use cache' directive, configuring cache lifetimes with cacheLife(), tagging cached data with cacheTag(), invalidating caches with updateTag()/revalidateTag(), optimizing static vs dynamic content boundaries, debugging cache issues, and reviewing Cache Component implementations.
Guides building MCP servers enabling LLMs to interact with external services via tools. Covers best practices, TypeScript/Node (MCP SDK), Python (FastMCP).
Share bugs, ideas, or general feedback.
This is an execution-aware skill — check skills/log-now.learnings.md in the vault before proceeding. If it exists, apply accumulated learnings.
The user has invoked /log-now. Log the current session segment immediately and write the narrative pieces inline.
Unlike the hook-driven logs — which can only do the mechanical half because hooks can't reach MCPs — /log-now runs in an active model turn. You do both halves: run the shell script, then write the narrative.
Run the session-log shell script in manual mode:
TRANSCRIPT="$(find ~/.claude/projects -name '*.jsonl' -print0 2>/dev/null | xargs -0 ls -t 2>/dev/null | head -1)"
SESSION_ID="$(basename "$TRANSCRIPT" .jsonl)"
WORKBENCH_LOG_MODE=manual bash "${CLAUDE_PLUGIN_ROOT}/hooks/session-log.sh" <<EOF
{
"session_id": "$SESSION_ID",
"transcript_path": "$TRANSCRIPT",
"hook_event_name": "ManualLogNow"
}
EOF
If the heuristic fails, ask the user for the transcript path.
After the script runs, read ~/.claude-memory-cache/pending-summaries/<session_id>.json to find the log path.
Read the raw log. Based on the log contents AND your own lived memory of this session, write the summary.
Read ${CLAUDE_PLUGIN_ROOT}/references/summary-format.md for the required shape. Set mode: manual. Write via mcp__plugin_workbench-core_memory__write.
Because you're in-session, your summary should be richer than what the auto summary-writer produces — you have context the raw JSONL doesn't capture.
Read ${CLAUDE_PLUGIN_ROOT}/references/decision-promotion.md for criteria.
Read ${CLAUDE_PLUGIN_ROOT}/references/vault-conventions.md for conventions.
Delete the pending-summary marker:
rm ~/.claude-memory-cache/pending-summaries/<session_id>.json
Tell the user what you wrote: log path, summary path, any decisions promoted. Keep it terse.
pending-summaries/ directory — only your marker.