Agent skill

Kmp Starter Memory

by DevAtrii in DevAtrii/Kmp-Starter-Template

Persistent per-project memory for KMP Starter Template apps.

MITAuto-check passedDevelopment

Install Kmp Starter Memory

skills CLI
$ npx skills add DevAtrii/Kmp-Starter-Template --skill kmp-starter-memory -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install DevAtrii/Kmp-Starter-Template kmp-starter-memory --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/DevAtrii/Kmp-Starter-Template.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/kmp-starter/memory .claude/skills/kmp-starter-memory && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
kmp-starter-memory
GitHub stars
167
Token cost
~963 tokens
SKILL.md length
364 words
Files
1
Skills in repo
23
Repo updated
First seen
Licence
MIT

At a glance

Persistent per-project memory for KMP Starter Template apps.

  • Works in 3 steps: Compute the {project} slug. → If… → Read structure.md / decisions.md /…
  • Starting work on a project (read memory first)
  • SKILL.md covers Storage location, When to read, What to store in memory.md and What to store in structure.md, plus 3 more sections
  • Calls python3

What it does

Kmp Starter Memory is an agent skill from DevAtrii/Kmp-Starter-Template. Persistent per-project memory for KMP Starter Template apps. Use when starting work on a project (read memory first), when learning the project structure, or after making significant changes (write/update memory).

Its SKILL.md is about 960 tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Development, covering Project scaffolding and Agent memory. The repository describes itself as: Starter template is a multi-module KMP boilerplate for Android & iOS with clean architecture and high-quality code. It handles basics like remote config, in-app purchases, and… The licence is MIT.

When your agent uses it

  • Starting work on a project (read memory first)
  • Learning the project structure
  • After making significant changes (write/update memory)

Example prompts

  • “/kmp-starter-memory”

Requirements

  • Python 3

Workflow steps

3 steps, taken from the first numbered list in SKILL.md.

  1. Compute the {project} slug.
  2. If {skill}/.skill-storage/{project}/memory.md exists, read it before exploring code.
  3. Read structure.md / decisions.md / progress.md only when relevant to the task.

What it can do on your machine

Read from SKILL.md and the folder at commit a8f7b27. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • python3

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Kmp Starter Memory loads about 963 tokens when it runs. Until then it costs about 58 tokens; SKILL.md has 364 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~58
When it runs · the whole SKILL.md, loaded when a task matches
~963

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from DevAtrii/Kmp-Starter-Template at commit a8f7b27, republished under its MIT licence (© DevAtrii). 364 words, ~963 tokens.

Download SKILL.mdSave it as .claude/skills/kmp-starter-memory/SKILL.md (or your agent's skills folder).
name
kmp-starter-memory
description
Persistent per-project memory for KMP Starter Template apps. Use when starting work on a project (read memory first), when learning the project structure, or after making significant changes (write/update memory).
author
DevAtrii
license
MIT

Memory

Keep a lightweight, persistent memory of each project so future sessions can resume without re-exploring everything.

Storage location

Store memory inside the skill directory, keyed by project:

{skill}/.skill-storage/{project}/
    memory.md        # REQUIRED — core memory (see below)
    structure.md     # optional — module / feature / screen layout
    decisions.md     # optional — key decisions made + why
    progress.md      # optional — current status, blockers, next steps
  • {skill} = the directory containing this SKILL.md.
  • {project} = a stable, filesystem-safe slug for the project (lowercase, hyphens). Derive from the app/package name (e.g. com.example.myapp → myapp, or the app name notes-app). Use the same slug every session for the same project.
Generate structure.md automatically

Use the bundled scan-project.py to produce structure.md (module tree, feature slices, nav routes, Koin wiring, versions):

bash
python3 scan-project.py --root /path/to/project --project my-app

It writes {skill}/.skill-storage/{project}/structure.md directly. Run it at the start of a session on an unfamiliar project, then read the file instead of re-walking the whole tree.

When to read

At the start of any session on a project:

  1. Compute the {project} slug.
  2. If {skill}/.skill-storage/{project}/memory.md exists, read it before exploring code.
  3. Read structure.md / decisions.md / progress.md only when relevant to the task.

If memory does not exist, create it after "Phase 1 — Understand the project".

What to store in memory.md

Keep it concise and factual. Use this template:

markdown
# Project Memory — {App Name}

## What it is
[One or two sentences describing the app and its purpose.]

## Goal
[What the app is trying to achieve / deliver.]

## Target audience
[Who it is for.]

## Special requirements
- [Any non-default constraints, integrations, backend, privacy, platforms, languages.]

## Key facts
- Package: `{your-package}`
- Features enabled: [database / purchases / analytics / remote config / notifications / locale ...]
- Min SDK / iOS target: ...

## Current state
[What is built, what is in progress, known gaps.]

Add/update sections only as needed. Do not dump entire file trees; summarize.

What to store in structure.md

A high-level map of the code you actually depend on:

  • Module layout (settings.gradle.kts includes, notable features/*).
  • Which screens exist and their route classes (e.g. AppScreens.Home).
  • Which feature slices exist and where (data / domain / presentation).
  • Any custom infrastructure that was added (not part of Starter).
Show full SKILL.md (128 more words)Show less

What to store in decisions.md

One line per non-obvious decision: what was chosen and why (e.g. "used StarterFileManager cache instead of Downloads because files are temporary", "did not add a Logic for X because it's a one-line repository call").

When to write / update

Write or update memory when:

  • A new feature, screen, or module is added/removed.
  • A significant architectural decision is made.
  • Project metadata changes (package, enabled features).
  • The task ends with unfinished work (record it in progress.md).

Keep updates small and additive. Prefer editing an existing file over creating many files.

Rules

  • Never store secrets, API keys, or tokens in memory files.
  • Keep {project} slug stable across sessions.
  • Memory is a convenience, not a source of truth — always verify against the actual codebase when a decision matters.

© DevAtrii, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .agents/skills/kmp-starter/memory of DevAtrii/Kmp-Starter-Template.

Open the folder on GitHubat commit a8f7b27

Compare with similar skills

Kmp Starter Memory next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Kmp Starter Memory compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Kmp Starter Memory this skillDevAtrii/Kmp-Starter-Template167—~963Automated safety check: PassMIT
Agent App CreatorCraftOS-dev/CraftBot392—~4.8kAutomated safety check: PassMIT
Satorisundial-org/awesome-openclaw-skills663—~1.4kAutomated safety check: PassNone
Project Docscoco-research/coco482—~3.6kAutomated safety check: PassCustom licence
Intake ProjectAperivue/medsci-skills329—~1.6kAutomated safety check: PassMIT
Ax Agent Contextdosco/aithy107—~903Automated safety check: PassApache-2.0

Similar skills

  • Agent App Creator

    CraftOS-dev/CraftBot

    Create Agent App applications (PocketBase backend, React kit frontend).

    392 GitHub stars~4.8k tokensUpdated yesterday
    DevelopmentAuto-check passed
  • Satori

    sundial-org/awesome-openclaw-skills

    Persistent long term memory for for continuity in ai sessions between providers and codegen tools.

    663 GitHub stars~1.4k tokensUpdated 7 mo ago
    DevelopmentAuto-check passed
  • Project Docs

    coco-research/coco

    A skill your agent uses when starting a new project, onboarding to an existing product, or auditing an existing folder structure against the standard.

    482 GitHub stars~3.6k tokensUpdated today
    Productivity & AutomationAuto-check passed
  • Intake Project

    Aperivue/medsci-skills

    A skill your agent uses when starting or inheriting a radiology research project and need to know what it is.

    329 GitHub stars~1.6k tokensUpdated 2 days ago
    Research & ScienceAuto-check passed
  • Ax Agent Context

    dosco/aithy

    This skill helps an LLM pick the right AxAgent context tool for a job - contextMap for recurring corpora, contextPolicy presets for within-run trajectory compaction, agent.optimize for offline GEPA…

    107 GitHub stars~903 tokensUpdated 1 mo ago
    Agent WorkflowsAuto-check passed
  • Commit Context Lookup

    rohitg00/agentmemory

    Traces a file, function or line back to the agent session behind its current commit, using git blame and a memory lookup, and reports only what the records show.

    29k GitHub stars~522 tokensUpdated yesterday
    DevelopmentAuto-check passed

More from DevAtrii/Kmp-Starter-Template

All 23 skills in this repo
  • Kmp Starter

    DevAtrii/Kmp-Starter-Template

    Build production Android & iOS apps on the KMP Starter Template (Clean Architecture, MVI, Koin, Compose Multiplatform).

    167 GitHub stars~3.3k tokensUpdated 1 mo ago
    Auto-check passed
  • Kmp Starter Architecture

    DevAtrii/Kmp-Starter-Template

    Clean Architecture module and package layout for KMP Starter Template features — data/domain/presentation layers, core dependency rule, screen-based package organization, the Screen vs Content…

    167 GitHub stars~1.6k tokensUpdated 1 mo ago
    Auto-check passed
  • Kmp Starter Create Fresh App

    DevAtrii/Kmp-Starter-Template

    Research and plan a brand-new app before scaffolding it on the KMP Starter Template.

    167 GitHub stars~1.4k tokensUpdated 1 mo ago
    Auto-check passed
  • Kmp Starter Data

    DevAtrii/Kmp-Starter-Template

    Data layer on the KMP Starter Template — repositories and data sources, Logics guidance, DataStore persistence, Room database, StarterFileManager, and the Calf file picker.

    167 GitHub stars~1.1k tokensUpdated 1 mo ago
    Auto-check passed
  • Kmp Starter Feature Analytics

    DevAtrii/Kmp-Starter-Template

    The KMP Starter Template analytics system — AppEvent/EventsTracker, Analytics routing, Mixpanel + Firebase providers, combining providers, and runtime swaps.

    167 GitHub stars~1.8k tokensUpdated 1 mo ago
    Auto-check passed
  • Kmp Starter Feature Database

    DevAtrii/Kmp-Starter-Template

    The KMP Starter Template Room database — entities, DAOs, migrations, and DB configuration in features/database.

    167 GitHub stars~806 tokensUpdated 1 mo ago
    Auto-check passed

Questions about Kmp Starter Memory

What does Kmp Starter Memory do?

Persistent per-project memory for KMP Starter Template apps. Kmp Starter Memory is an agent skill from DevAtrii/Kmp-Starter-Template. Persistent per-project memory for KMP Starter Template apps.

When should I use Kmp Starter Memory?

Kmp Starter Memory fits situations like: starting work on a project (read memory first); learning the project structure; after making significant changes (write/update memory).

How do I install Kmp Starter Memory in Claude Code?

Run `npx skills add DevAtrii/Kmp-Starter-Template --skill kmp-starter-memory -a claude-code`. Or copy the skill folder (.agents/skills/kmp-starter/memory in DevAtrii/Kmp-Starter-Template) into .claude/skills/kmp-starter-memory in your project. Claude Code loads it when a task matches its description.

How do I install Kmp Starter Memory in Codex?

Run `npx skills add DevAtrii/Kmp-Starter-Template --skill kmp-starter-memory -a codex`. Or copy the skill folder (.agents/skills/kmp-starter/memory in DevAtrii/Kmp-Starter-Template) into .agents/skills/kmp-starter-memory in your project. Codex loads it when a task matches its description.

Can I use Kmp Starter Memory in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add DevAtrii/Kmp-Starter-Template --skill kmp-starter-memory -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/kmp-starter-memory, .gemini/skills/kmp-starter-memory, .github/skills/kmp-starter-memory and .opencode/skills/kmp-starter-memory in your project.

What does Kmp Starter Memory need to run?

Going by SKILL.md and its folder, Kmp Starter Memory needs the command-line tools its instructions call (python3). Our summary lists: Python 3.

Does Kmp Starter Memory access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Kmp Starter Memory safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Kmp Starter Memory use?

Kmp Starter Memory is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Kmp Starter Memory use?

About 963 tokens (SKILL.md is roughly 3.9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Kmp Starter Memory?

Skills that share tags, products or a category with Kmp Starter Memory: Agent App Creator (CraftOS-dev/CraftBot, 392 stars), Satori (sundial-org/awesome-openclaw-skills, 663 stars), Project Docs (coco-research/coco, 482 stars) and Intake Project (Aperivue/medsci-skills, 329 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Kmp Starter Memory?

DevAtrii (a GitHub user) maintains it in DevAtrii/Kmp-Starter-Template, which has 167 GitHub stars. The repository holds 23 skills in this directory. The repository was last updated on September 1, 2026.

Source: DevAtrii/Kmp-Starter-Template on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.