Session wrap-up that reads the breadcrumb trail your toolkit already left — not a cold reconstruction. Multi-repo aware: surfaces what shipped across every repo the session touched, what's uncommitted, and what's unpushed, then gates commit + push on the current repo only (read wide, mutate narrow).
This skill should be used when the user says `/vibe-wrap:evolve-wrap` and wants vibe-wrap to reflect on its own past sessions and propose plugin improvements to itself. Pattern
Internal SKILL — not a slash command. Append-only friction capture for vibe-wrap (Pattern
Internal SKILL — not a slash command. Shared behavior for vibe-wrap commands — voice rules (sentence-case headings, em-dashes, no emoji, no corporate speak), bumper-lanes invariant (default no-action at every gate, every gate has a skip path), persona adaptation (reads `shared.preferences.persona` from the unified builder profile and adapts wrap voice to professor / cohort / superdev / architect / coach), friction-trigger contract linking each command to its `friction-logger.log()` invocations. Referenced by every other vibe-wrap SKILL. See `references/voice.md`, `references/persona-adaptation.md`, `references/friction-triggers.md`.
Internal SKILL — not a user-invocable slash command. Sibling vibe plugins (or non-vibe tools that opt in) invoke `plant(source, command, phase, outcome=null, payload=null)` at command-start (or any point worth marking) to drop one breadcrumb line into the active session's breadcrumb file. No-op-safe — silent failure if vibe-wrap isn't installed or the session UUID can't be resolved (falls back to `_orphan.jsonl`). Writes one JSONL line to `~/.claude/plugins/data/vibe-wrap/breadcrumbs/<session-uuid>.jsonl`. Forward-compat — unknown payload fields are written verbatim. See `references/breadcrumb-contract.md` for the full schema and contract for sibling plugin authors.
Internal SKILL — not a slash command. Two-phase append-only session log for vibe-wrap's OWN sessions: a sentinel entry at command start (`outcome=in_progress`) and a terminal entry at command end, paired by sessionUUID. Distinct from breadcrumb capture — this is vibe-wrap's self-evolution instrumentation (Pattern #2), not the cross-plugin trail. Storage: `~/.claude/plugins/data/vibe-wrap/sessions/<YYYY-MM-DD>.jsonl`. Pattern #11 namespace isolation — writes ONLY inside vibe-wrap's data dir. Invoked by every vibe-wrap command at start and end.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Sessions wrap themselves when the trail is already there.
End-of-session wrap-up for Claude Code — it reads the breadcrumb trail your toolkit already left instead of cold-reconstructing from git log and memory. One gesture at session end gives you a clean handoff: what shipped, what's still uncommitted, what's still unpushed, and a written record you can read, share, or paste into a status update.
It's not a prompt for the next agent. It's a "what we shipped today" summary.
End-of-session reconstruction is expensive when work hasn't been committed in steady passes. Vibe Wrap solves that with a breadcrumb pattern — sibling vibe plugins already write structured session-logs, friction, and wins as part of their own self-evolution loop, so Vibe Wrap reads that trail at session end. Zero coupling, zero extra instrumentation required.
/vibe-wrap — the main end-of-session wrap. Reads the trail + git state + your decision log, renders a six-section summary to docs/session-wraps/<ts>.md and inline, then runs interactive commit / push / decision-log / dashboard-bridge gates./vibe-wrap:status — read-only, mid-session: what trail's been picked up so far./vibe-wrap:plant — internal; siblings (or any tool) drop a richer breadcrumb at command-start. No-op-safe./vibe-wrap:evolve-wrap — self-evolution (Pattern #1): reads Vibe Wrap's own session + friction logs and proposes plugin improvements. Never auto-applies.Every gate defaults to no-action. Nothing commits, pushes, or writes without an explicit yes — and secret-pattern matches (.env*, *.pem, *.key, *token*) require a second confirmation. The wrap surfaces state; you decide what happens to it.
Vibe Wrap reads and (optionally) appends to a decision log. Four backends, chosen on first run or via config: Markdown file (default), JSONL file, 626Labs MCP (when reachable), or disabled. It's a marketplace plugin — useful for any builder, not just 626Labs. The 626Labs dashboard bridge is an MCP-only, threshold-gated extra.
Canary (this repo, latest main):
/plugin marketplace add estevanhernandez-stack-ed/vibe-wrap
/plugin install vibe-wrap
Stable ships via the aggregated 626Labs marketplace once a tagged release is pinned there.
PATH. Pure stdlib — no third-party dependencies.Want your tool to enrich the trail? See the breadcrumb contract at plugins/vibe-wrap/skills/wrap/references/breadcrumb-contract.md. Vibe Wrap already reads any plugin that writes the ~/.claude/plugins/data/<name>/sessions/ + friction.jsonl convention — :plant is the opt-in for richer attribution.
MIT © 626Labs LLC
npx claudepluginhub estevanhernandez-stack-ed/vibe-wrap --plugin vibe-wrapAI-powered documentation gap analyzer. Scans your codebase, classifies your project, identifies missing technical documentation, and generates professional docs from your existing artifacts.
Vibe Cartographer — plot your course from idea to shipped app. Spec-driven development delivered as eight slash commands: onboard, scope, prd, spec, checklist, build, iterate, reflect.
Test analyzer and generator for vibe-coded apps. Classifies your app by type and maturity tier, audits existing coverage honestly (no cherry-picked denominators), and generates the tests that actually matter proportional to deployment risk.
626 Labs design system + publishing umbrella as Claude Code skills. Design skill provides brand tokens, type, UI kit, and editorial layer for any 626 surface. Publishing skill orients Claude to the studios/works/library/packages bucket model and routes to the right Lead Writer persona.
Post-ship product iteration — pick a mode (feature-add, competitive, ux-polish, bug-bash), ship one PR. Ingests competitor changelogs, framework releases, and your feedback.md. Maintains your Atlas. Sibling to vibe-cartographer.
Feature development with code-architect/explorer/reviewer agents, CLAUDE.md audit and session learnings, and Agent Skills creation with eval benchmarking from Anthropic.
Production-grade engineering skills for AI coding agents — covering the full software development lifecycle from spec to ship.
Permanent coding companion for Claude Code — survives any update. MCP-based terminal pet with ASCII art, stats, reactions, and personality.
Claude harness - A harness for solo developers (Vibecoders) to handle full-cycle contract development.
Complete Claude Code configs for mobile development - Android, iOS, and Kotlin Multiplatform