Skip to main content
skill CX Codex

groove-utilities-memory-promises — Claude Code skill

Captures deferred tasks during coding sessions and tracks them as items in the beans backend for later review.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add andreadellacorte/groove --skill groove-utilities-memory-promises -a claude-code
Open on GitHub

What groove-utilities-memory-promises does

This skill catches conversational items you plan to revisit later, preventing them from vanishing into chat history. It routes promises into a structured hierarchy in beans, placing them under a dedicated Groove Memory milestone and a Promises epic. Tasks get flagged with a deferred status so they stay separated from your active priorities.

Reviewing and updating the list takes minimal input. Passing --list prints open entries in an ordered index, while --resolve <N> closes a specific record by number. The agent will not capture items automatically without asking for explicit user confirmation first.

If .groove/index.md sets tasks.backend to none, the skill halts execution immediately and instructs you to run /groove-utilities-task-install.

What it helps with

  • record deferred thoughts without disrupting current session flow
  • list pending session promises in a numbered overview
  • close an unresolved promise by its index number

Example prompts for the agent

  • Run groove-utilities-memory-promises to record that we need to write unit tests for the auth module
  • Show all pending items with groove-utilities-memory-promises --list
  • Mark promise 2 as completed using groove-utilities-memory-promises --resolve 2

How to install groove-utilities-memory-promises

1Skills CLI (Vercel)

npx skills add andreadellacorte/groove --skill groove-utilities-memory-promises -a claude-code

2Manually: put the files in place

~/.claude/skills/groove-utilities-memory-promises/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/groove-utilities-memory-promises/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about groove-utilities-memory-promises

How to install groove-utilities-memory-promises in Claude Code?

Run npx skills add andreadellacorte/groove --skill groove-utilities-memory-promises -a claude-code in your terminal.

What does groove-utilities-memory-promises do?

It files deferred session tasks into a dedicated beans task epic and lets you inspect or resolve them via command flags.

What dependencies does this skill require?

It depends on the beans CLI tool and an active task backend configured in .groove/index.md.

Source text

SKILL.md · MIT
---
name: groove-utilities-memory-promises
description: "Capture and resolve deferred items from a session ('we'll come back to that'). Use $ARGUMENTS as the promise text, or --list / --resolve N."
license: MIT
allowed-tools: Read Write Edit Glob Bash(beans:*) AskUserQuestion
metadata:
  author: andreadellacorte
---

# groove-utilities-memory-promises

Use $ARGUMENTS as the promise text if provided, or a flag:
- `--list`: show all open promises
- `--resolve <N>`: mark promise N as resolved

## Outcome

Deferred items from a session are captured before they are forgotten, or existing open promises are reviewed and resolved.

## Acceptance Criteria

- New promise captured with date and optional context
- `--list` shows all open promises clearly numbered
- `--resolve N` marks promise N as resolved

## Task backend

Read `tasks.backend` from `.groove/index.md`. Promises are tracked as tasks in the configured backend. If no task backend is configured (`tasks.backend: none`), tell the user to run `/groove-utilities-task-install` first.

Promises are stored as tasks under a shared "Groove Memory" milestone → "Promises" epic hierarchy. This keeps them organised and out of the main work task list.

### Ensure parent hierarchy

Before any operation, resolve or create the parent epic:

1. Find or create the **Groove Memory** milestone:
   - `beans list -t milestone --search "Groove Memory" -q` — if output is non-empty, use the first ID; otherwise `beans create "Groove Memory" -t milestone` and capture the new ID
2. Find or create the **Promises** epic under that milestone:
   - `beans list -t epic --parent <milestone-id> --search "Promises" -q` — if non-empty, use first ID; otherwise `beans create "Promises" -t epic --parent <milestone-id>` and capture the new ID

Use the Promises epic ID as `<parent-id>` for all promise tasks.

### `--list`

1. Resolve `<parent-id>` (see above)
2. `beans list --parent <parent-id> -s todo -t task`
3. Display as numbered list: `1. [<id>] <title>` — position number for `--resolve N`, beans ID in brackets
4. If empty: print "No open promises."

### `--resolve <N>`

1. Resolve `<parent-id>`; run `beans list --parent <parent-id> -s todo -t task`
2. Find the Nth item; if not found: print "No open promise #N" and exit
3. `beans update <id> -s completed`
4. Confirm: "Promise #N resolved."

### Default — add a promise

1. Resolve `<parent-id>` (see above)
2. Get promise text from $ARGUMENTS if provided; otherwise ask: "What's being deferred?"
3. Optionally ask: "Any context? (enter to skip)"
4. `beans create "<text>" -t task --parent <parent-id> -s todo -p deferred`
5. If context given: `beans update <id> -d "<context>"`
6. Confirm: "Promise captured: [<id>] <text>"

## Constraints

- Read `tasks.backend` from `.groove/index.md` at the start of every invocation
- Requires a configured task backend — if `tasks.backend: none`, prompt user to install one
- Never auto-capture promises without user confirmation — always explicit
- Milestone/epic parent hierarchy is idempotent — always check before creating; never create duplicates
- Use `-s todo` for open, `-s completed` for resolved; `-p deferred` to signal not active work
- Do not resolve all promises automatically; resolve one at a time unless user asks for `--resolve-all`
- `--resolve-all` requires confirmation: "Resolve all N open promises?" before proceeding

© andreadellacorte · MIT · andreadellacorte/groove

Similar skills

All skills for Agent setup: memory, context, orchestration →

From the same set41

groove-utilities-task-archive Archives completed and scrapped tasks in the configured backend. Use it when you want to close a batch of finished work with a confirmation step. CX Codex +1 532installs groove-admin-claude-hooks Installs groove’s native shell hooks in .claude/settings.json. They track git activity, prepare session-capture drafts, and guard managed paths. CX Codex +1 530installs groove-utilities-memory-mistakes This skill turns an agent mistake into a tracked incident, fixes its cause, and records the lesson in project memory. Use it to stop repeat errors. CX Codex +1 529installs groove-admin-update Pulls the latest groove skills and applies pending migrations to your local groove state. Useful when a new release lands but your config and memory are still on the old one. CX Codex +1 528installs groove-work-doc Creates a reference document explaining how a codebase component or concept works. Use it when future agents or contributors need concrete project context. CX Codex +1 518installs groove-admin-cursor-hooks Registers groove's native Cursor hooks in .cursor/hooks.json: context re-priming after compaction, end-of-day reminders, a git commit buffer and managed-path protection. CX Codex +1 518installs groove-utilities-memory-retrospective Summarizes workflow trends from .groove/memory/ by analyzing session ratings, mistake patterns, and recorded learnings. Helpful for weekly or monthly development check-ins. CX Codex +1 516installs groove-utilities-memory-graduate Promotes a stable workflow insight from .groove/memory/learned/ into AGENTS.md, so the rule applies to every future session without running prime. CX Codex +1 510installs

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor