recall

Reconstruct recent working context from chat history and shared records into a status brief.

Updated Sep 2, 2026
One-click install
npx skills add https://github.com/jnyross/pstack-muse --skill recall-jnyross
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: recall
Source: https://github.com/jnyross/pstack-muse/tree/main/skills/recall
Command: npx skills add https://github.com/jnyross/pstack-muse --skill recall-jnyross

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When you return to a project after days away, you lose track of what you were doing, which PRs shipped, which fixes got reverted, and what users are still reporting. This Skill rebuilds that working context from your own chat transcripts and the shared engineering record, then hands back a tight current-state brief so you can resume work immediately. ## Core Features & Use Cases - Parallel transcript mining: Fans out subagents across your recent chat history to extract topics, decisions, open threads, and artifacts, each cited by chat UUID. - Shared-record sweep: Searches source control, issue trackers, chat channels, docs, and error tracking for the history around a named feature or bug, including reverted fixes and recurring user reports. - Live-state verification: Checks surfaced PRs, branches, and tickets against git and GitHub so the brief reflects current truth, not stale history. - Use Case: You type "catch me up on the payments retry work" after a week off, and receive a capsule with thread statuses like [merged #412] and [reverted #398], the top recurring problems, and the single most useful next action. ## Quick Start Ask the assistant to recall your recent work on a specific feature and produce a current-state brief with thread statuses and the next move.

Frequently Asked Questions about recall

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I catch up on my recent work after time away?

Ask for a recall of your work on a topic, optionally naming a time window. The skill mines your recent chat transcripts in parallel, sweeps the shared record for the named feature, verifies PRs and branches against live git state, and returns a brief with thread statuses and a next move.

How does chat history search work across sessions?

Parallel subagents each take a slice of the transcript corpus, order candidates by real modification time, grep for the topic, and read only matching regions. Each returns a structured summary citing the chat UUID, so the main thread keeps only findings, not raw transcripts.

Does recall search more than my own chat history?

Yes. When the topic names a feature, file, or bug, it also sweeps the shared record: source control, issue trackers, chat channels, docs, and error tracking. This surfaces reverted fixes and recurring user reports that your own transcripts never captured.

When should I use session pickup instead of full recall?

Use session pickup when you want to resume one specific prior chat. Recall is for rebuilding working context across many recent chats before starting or resuming work, and pure activity questions like "what did I do this week" skip the shared-record sweep entirely.

Why might a recall brief contain stale information?

Transcripts and old tickets are history, not current truth. The skill mitigates this by verifying surfaced PRs, branches, and tickets with git and GitHub before writing the brief, and by reading full transcripts when the answer hinges on what an agent actually did.