By JoaoAnt42
A Claude Code way-of-work cycle. Brainstorm → plan → auto-review → human-review → workbench → TDD-red → implement → smoke → final-review → receiving-review. Independent agents for test/impl/review.
Entry point for the Memento way-of-work. Use when the user invokes `/use_new_superpowers`, `/memento`, or starts any multi-step coding task that should go through the Memento cycle. Sizes the task (small or large) and routes it. Replaces the superpowers:using-superpowers flow.
Step 1 of Memento. Use when receiving a new task (markdown, screenshots, verbal description) that will go through the Memento cycle, before any planning or code.
Step 2 of Memento. Use after brainstorming converges, to write a durable plan markdown in the Obsidian plans directory. Enforces propose-4-options questioning and keeps plans out of the repo.
Step 3 of Memento. Use after a plan is drafted, before human review. Runs a multi-turn discussion between Devil's Advocate, Simplifier, and Orchestrator subagents to stress-test and rewrite the plan in place.
Step 4 of Memento. Use after auto-review. Submits the updated plan to the user for approval, rejection, or revision. Loops back to planning on rejection.
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.
A Claude Code way-of-work cycle. Maximize confidence over speed — and match process weight to task size.
Every task starts with a prior-art sweep (find any existing implementation before scoping), then is sized Small or Large:
Independent agents for test / implement / review. Red-commit SHA is the handoff artifact.
/plugin marketplace add JoaoAnt42/memento
/plugin install memento@memento
That's it — skills and the /use_memento command are now available.
git clone https://github.com/JoaoAnt42/memento
cd memento
for d in skills/*/; do
name="$(basename "$d")"
mkdir -p ~/.claude/skills/"$name"
cp "$d/SKILL.md" ~/.claude/skills/"$name"/SKILL.md
done
cp commands/use_memento.md ~/.claude/commands/use_memento.md
The plugin installs from a version-pinned cache — update-all (/plugin marketplace update) only re-pulls when the version field changes, not on every commit. To ship an edit:
version in both .claude-plugin/plugin.json and .claude-plugin/marketplace.json (all three fields must match).master./plugin marketplace update memento then /plugin update memento@memento.Skip the version bump and the update is a silent no-op — Claude keeps serving the cached version. Pushing requires the JoaoAnt42 GitHub identity (the clone's origin uses SSH).
/use_memento <task description>
Entry point is memento-0-using. Each step invokes the next skill in sequence.
| Step | Skill | Purpose |
|---|---|---|
| 0 | memento-0-using | Entry point. Prior-art sweep, then size Small/Large and route |
| 1 | memento-1-brainstorming | Explore intent, self-research first |
| 2 | memento-2-planning | Write plan in Obsidian, propose-4-options |
| 3 | memento-3-auto-review | Devil's Advocate + Simplifier + Orchestrator |
| 4 | memento-4-human-review | Submit plan; on approve, create a git worktree per repo; rejection loops back to step 2 |
| 6 | memento-6-tdd-red | Write failing tests, commit red SHA |
| 7 | memento-7-implementing | Independent impl agent, input = red SHA |
| 7.5 | memento-7b-human-smoke | Optional; gated by needs_human_smoke: true |
| 8 | memento-8-final-review | Parallel reviewers, precedence: Bugs > CRAP > Simplifier > DA > Tests |
| 8.5 | memento-8b-cross-pr-review | Optional, multi-repo only (repos: ≥2). Cross-PR contract check; applies fixes in worktrees and pushes |
| 9 | memento-9-receiving-review | Verify, push back, or implement. Loops back to step 8 on non-trivial review changes. |
npx claudepluginhub joaoant42/memento --plugin mementoHarness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Persistent file-based planning for AI coding agents. Crash-proof markdown plans (task_plan.md, findings.md, progress.md) that survive context loss and /clear, with an opt-in completion gate and multi-agent shared state. Manus-style. Works with Claude Code, Codex CLI, Cursor, Kiro, OpenCode and 60+ agents via the SKILL.md standard. Includes Arabic, German, Spanish, and Chinese (Simplified and Traditional).
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.