><(((*>

Just Keep Swimming

Context-aware handoffs for Claude Code. Your AI agent never loses track of what it's building — no matter how many sessions it takes.

Install in 10 seconds ↓

Context rot is silent and expensive

When an AI agent's context window fills up, the system compresses earlier messages. Nuanced details vanish. The agent keeps working — but now it's making subtly wrong decisions.

🔄

Repeated questions

The agent re-asks things it already resolved an hour ago, wasting time and breaking flow.

🚧

Architectural drift

Implementations silently deviate from decisions made earlier in the session when those decisions get compressed away.

💥

Regression by amnesia

The agent breaks things it just built because it forgot why they were built that way.

Write it down before you forget

Just Keep Swimming makes the agent document everything it knows before context degrades — automatically, with zero ceremony.

1

Plan

Create or import an implementation plan

2

Execute

Work through phases continuously

3

Handoff

Capture full context before it fades

4

Resume

Next session picks up seamlessly

Everything you need. Nothing you don't.

Core

Automatic handoffs

The agent monitors its own context health and creates handoff documents automatically when things get heavy. No manual intervention needed.

Smart

Cumulative context

Each handoff condenses ALL prior sessions' knowledge. The next agent reads one document — not five — and has everything.

Flexible

Two execution modes

Autonomous (just keep swimming) or interactive (pause for review). Switch with a single flag.

Precise

File:line references

Handoffs include exact file paths and line numbers. No vague summaries — the next agent knows exactly where to look.

Tracked

Blockers & amendments

Blockers are surfaced on resume. Plan amendments capture where reality diverged. Nothing falls through the cracks.

Light

One plugin, zero config

Install with a single command. No state files, no setup wizards. Plans and handoffs are just markdown in your repo.

Autonomous

Night build pipeline

End-of-day command that reviews commits, code reviews, builds, deploys, writes a summary, and pushes to git. Fully unattended.

Adaptive

Context-aware budgets

Detects your context window size (200k vs 1M) and sets hard phase limits. No "just one more thing" — it hands off before context degrades.

Instant

Plan from conversation

Already discussing architecture? Use --from-context to distill the conversation into a structured plan. Add --now to skip confirmations.

Durable

Post-delivery maintenance

Completed plans stay useful. Bug reports get structured debugging with maintenance logs, not ad-hoc guessing.

What a handoff actually looks like

Real context preservation, not a vague summary. This is what the next session reads to restore full context.

  2026-03-12-handoff-001.md
# Handoff — API Migration — Session 001 **Date**: 2026-03-12 14:30 **Progress**: 2 of 6 phases complete (~33%) ## Completed This Session - [x] Phase 0: Extract database queries — `services/database/_core.js:45-120` - [x] Phase 1: Create service interfaces — `services/api/userService.js:1-85` ## Remaining Work - [ ] Phase 2: Route migration (12 route files) - [ ] Phase 3: Cache layer separation - [ ] Phase 4: WebSocket event routing - [ ] Phase 5: Integration testing ## Blockers & Open Questions - Blocker: Task deadline queries use nested Firebase paths that don't support inequality filters — need to restructure or use client-side filtering ## Key Learnings (This Session) - Redis cache keys follow `conv:{userId}:{convId}` scheme - Users cache has 60-min TTL with stampede protection - `_trackedRead()` is mandatory for all Firebase reads - Database service is 700KB+ — splitting by collection is the right granularity ## Next Session 1. Read first: `routes/messages.js:200-350` (heaviest route) 2. Focus: Phase 2 — migrate routes one at a time 3. Watch out for: circular deps in socketService

Your workflow, your way

Autonomous Default

The agent executes continuously, creating handoffs on its own. Come back to progress or a clean handoff.

/justkeepswimming:go my-feature

Interactive --interactive

Pauses after each phase. Review results, adjust the plan, decide when to continue or hand off.

/justkeepswimming:go my-feature --interactive

Lightweight by design

Just Keep Swimming Heavy workflow systems
Installation 1 plugin command 30+ files
State tracking Handoff docs only STATE.md, ROADMAP.md, CONTEXT.md, etc.
Concepts to learn Plan, Execute, Handoff Projects, Milestones, Phases, Waves, Audits
Agent spawning None Planners, Executors, Verifiers, Researchers
Setup time 10 seconds Plugin install + configuration
Dependencies None required Plugin ecosystem
Commands 2 (plan + night build) 10+ specialized agents

Agents that think, not just execute

Most AI agents fail the same way: they pattern-match to common fixes and try them one by one. Just Keep Swimming ships with six problem-solving principles injected into every session — changing how the agent approaches problems.

1

Diagnose before you prescribe

Gather facts before forming hypotheses. Every minute understanding the problem saves ten minutes of wrong-direction fixes.

Anti-pattern: "This looks like X, so let me try the usual fix."
Correct: "Let me verify what's actually happening first."
2

Trace the full chain

List every condition required for success and verify each one. The broken condition is always the one nobody checks because it's "too basic."

3

Check what's actually there

Run diagnostics with an open mind. Read the actual output — don't skim for confirmation of what you already believe.

4

Know the silent failures

Systems fail without useful errors all the time. They fall back, return generic messages, or silently skip the broken path. Find the quiet one.

5

Minimum effective intervention

Fix the root cause and only the root cause. The best fix is the smallest change. If you changed more than needed, you didn't understand the problem well enough.

6

Resist "the usual fix"

If the user has tried everything, the obvious solutions are exhausted. The answer lives in the layer nobody checked yet. Don't re-run the playbook.

"Don't be a solution-guesser. Be a condition-verifier."

Every condition required for success must be verified. The one you skip is the one that's broken. The protocol is customizable — add your team's debugging lessons to THINKING.md and every future session inherits them.

Up and running in 10 seconds

Install as a plugin or copy two files. No dependencies.

1 Add the marketplace

/plugin marketplace add cyber-qais/justkeepswimming

2 Install the plugin

/plugin install justkeepswimming

3 Use it

Open any Claude Code session and run:

/justkeepswimming:go my-feature

Alternative: Manual install

mkdir -p ~/.claude/commands/justkeepswimming
cp commands/justkeepswimming/*.md ~/.claude/commands/justkeepswimming/

Alternative: npm

npm install justkeepswimming
Works with any Claude Code project. Optionally pairs with the superpowers plugin for auto-generated plans.
Available on GitHub · npm