Recap
- 171 installs
- 411 repo stars
- Updated April 29, 2026
- sharpdeveye/maestro
recap is a Maestro v2.0.0 agent command skill that summarizes the most recent .maestro/sessions/ file and last five decisions.jsonl entries so developers can resume coding-agent work in under 30 seconds.
About
recap is a user-invocable Maestro command skill (version 2.0.0) in sharpdeveye/maestro for fast session recovery at the start of a new coding-agent conversation. It reads the most recent file in .maestro/sessions/ by date prefix and the last five entries from .maestro/decisions.jsonl, then outputs a concise ASCII summary led by a PICK UP HERE section with specific next commands and files. If no session files exist, recap falls back to the last ten decisions; if no data exists, it directs the user to run /capture. The skill is designed as a 30-second read—not a full session dump—and forbids vague next steps or assuming context from the current chat. It pairs with /capture as the recommended session-start command in Maestro's 25-command workflow toolkit and requires /teach-maestro context when none exists.
- Reads latest .maestro/sessions/ file plus last 5 decisions.jsonl entries
- Outputs PICK UP HERE section with specific Maestro commands, not vague tasks
- Designed as 30-second read; falls back to last 10 decisions if no session file
- Maestro v2.0.0 utility command in 25-command workflow toolkit
- Directs to /capture when no persisted session history exists
Recap by the numbers
- 171 all-time installs (skills.sh)
- +5 installs in the week ending Jul 27, 2026 (Skillselion tracking)
- Ranked #3,105 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
- Data as of Jul 27, 2026 (Skillselion catalog sync)
npx skills add https://github.com/sharpdeveye/maestro --skill recapAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 171 |
|---|---|
| repo stars | ★ 411 |
| Last updated | April 29, 2026 |
| Repository | sharpdeveye/maestro ↗ |
How do you resume an AI coding session from saved context?
Run /recap at session start to read the latest .maestro/sessions/ file and surface specific next commands from the PICK UP HERE section.
Who is it for?
Developers starting a new Maestro session who previously ran /capture and need a 30-second briefing on where work stopped.
Skip if: Users without any .maestro/sessions/ or decisions.jsonl data, or teams that rely solely on live chat history instead of persisted Maestro records.
When should I use this skill?
The user starts a new coding session and asks what happened last time, where to pick up, or invokes /recap explicitly.
What you get
Concise session recap with commands run, files changed, open issues, and a PICK UP HERE list of specific next Maestro commands.
- concise session recap
- actionable next-command list
By the numbers
- Reads last 5 decisions from .maestro/decisions.jsonl; falls back to last 10 without session files
- Maestro v2.0.0 recap command in 25-command workflow toolkit
- Targets 30-second concise recap with PICK UP HERE action list
Files
MANDATORY PREPARATION
Invoke /agent-workflow — it contains workflow principles, anti-patterns, and the Context Gathering Protocol. Follow the protocol before proceeding — if no workflow context exists yet, you MUST run /teach-maestro first.
---
Get back up to speed fast. Recap reads the most recent session file and recent decisions to tell you where you left off and what to do next.
Data Sources
1. `.maestro/sessions/` — read the most recent session file (by filename date) 2. `.maestro/decisions.jsonl` — read the last 5 decisions
If no session files exist, read the last 10 decisions and summarize those instead.
If no data exists at all, respond: "No session history found. Run `/capture` at the end of your next session to start tracking."
Output Format
┌─────────────────────────────────────┐
│ LAST SESSION RECAP │
├─────────────────────────────────────┤
│ Date: YYYY-MM-DD │
│ Commands: /diagnose → /fortify (2) │
│ Files: 3 changed, 1 created │
│ Outcome: Completed │
├─────────────────────────────────────┤
│ PICK UP HERE: │
│ 1. Run /guard — rate limiting │
│ 2. Run /evaluate — adversarial tests│
└─────────────────────────────────────┘Recap Rules
- Be concise — this is a 30-second read, not a report
- Lead with the "PICK UP HERE" section — that's what the user needs
- If the session ended with open issues, highlight them
- Reference specific files and commands, not vague summaries
Recap Checklist
- [ ] Most recent session identified
- [ ] Commands and outcomes summarized in one line each
- [ ] "Pick up here" section has specific, actionable next steps
- [ ] Open issues from last session surfaced
Recommended Next Step
After recapping, run the first command listed in "Pick Up Here" to continue where you left off.
NEVER:
- Dump the entire session file — summarize it
- Recap without checking decisions.jsonl for recent activity
- Give vague next steps ("continue working") — be specific
- Assume context from the current conversation — only use persisted data
Related skills
Setup Matt Pocock SkillsScaffold the per-repo configuration that Matt Pocock’s engineering agent skills rely on so they understand the issue tracker, triage labels, and domain documentation la495k196k
Lark AppsConnect Claude, Cursor or custom agents directly to Lark (Feishu) for messaging, document automation, approval workflows and enterprise data access.402k
CavemanSlash token usage by roughly 75% while keeping every technical detail intact when working with Claude Code, Cursor or similar agents.391k94.7k
Lark Skill MakerQuickly turn any Lark/Feishu OpenAPI call or multi-step workflow into a reusable agent skill with its own SKILL.md.386k16k
Running Claude Code Via Litellm CopilotRun Claude Code at a fraction of the cost by routing requests through LiteLLM to the GitHub Copilot Chat API.270k72
Codex PetGenerate a complete Codex Pet spritesheet and metadata from one reference image without needing an OpenAI key or Codex Pro.246k10How it compares
Pick Maestro recap for structured .maestro/ session recovery; use git log when you only need commit history without agent decision context.
FAQ
What data sources does Maestro recap read?
Maestro recap reads the most recent file in .maestro/sessions/ sorted by date prefix and the last five entries in .maestro/decisions.jsonl. Without session files, it summarizes the last ten decisions instead.
How long should a Maestro recap output be?
Maestro recap targets a 30-second read with a PICK UP HERE section listing specific next commands and files. It summarizes rather than dumping the full session markdown file.
What if no Maestro session history exists?
Maestro recap responds that no session history was found and instructs the user to run /capture at the end of the next session to start tracking decisions and session summaries.