Save Progress logo

Save Progress

Community
KimYx0207
save-progress

Save current project task progress to local task state file for cross-session continuity. Use when the user asks to save progress, uses an equivalent localized trigger phrase, the session is interrupted, or the user wants to resume work later. Writes to .claude/project-task-state.json so next session can load it via SessionStart hook.

Overview

PublisherKimYx0207
RepositoryMeta_Kim
Skill namesave-progress
Stars
284
Forks
67
Bundled files
Instructions only
LicenseApache-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by KimYx0207 on GitHub. Read the source before you install it.

Installation

Install the Save Progress AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/KimYx0207/Meta_Kim/tree/main/canonical/runtime-assets/claude/commands/save-progress
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/KimYx0207/Meta_Kim.git /tmp/Meta_Kim
mkdir -p .claude/skills
cp -r /tmp/Meta_Kim/canonical/runtime-assets/claude/commands/save-progress .claude/skills/save-progress
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Save Progress in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Save Progress on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Save Progress is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Save Progress

Saves current task progress to .claude/project-task-state.json in the project directory so the next session can automatically load it and resume where you left off.

What to Save

Before invoking this command, you should know:

  • Current task: What are you working on RIGHT NOW?
  • Completed tasks: What has been finished in this session? (Be specific, not "various things")
  • Remaining tasks: What still needs to be done?

How to Invoke

bash
# Detect the Python hook path — it lives in ~/.claude/hooks/
python "$HOME/.claude/hooks/mcp_memory_global.py" \
  --mode save \
  --task "Describe current task in one sentence" \
  --done "Task 1 completed" "Task 2 completed" \
  --remaining "Next task to do" "Another remaining task"

Or use the Claude Code hooks path from the user profile if HOME is not set:

cmd
python "%USERPROFILE%/.claude/hooks/mcp_memory_global.py" ^
  --mode save ^
  --task "..." ^
  --done "..." ^
  --remaining "..."

When to Save

Trigger this when:

  • User explicitly asks to save progress
  • Session is being interrupted (detected via hook or user signal)
  • A major milestone is completed
  • Before switching to a different project
  • End of a work session

What NOT to Save

  • Generic "worked on project" — be specific: "completed dependency license audit for 4 READMEs"
  • Session noise — only save meaningful task state
  • Duplicate entries — if nothing meaningful changed, don't overwrite

Output

The command returns JSON with:

  • saved: boolean
  • file: path to the state file
  • total_sessions: number of sessions recorded
  • total_completed: total completed tasks across all sessions
  • last_session_completed: how many tasks completed this session
  • last_session_remaining: remaining tasks as of this save

Show the user a brief summary after saving.

Example

User says: "save progress"

bash
python "$HOME/.claude/hooks/mcp_memory_global.py" \
  --mode save \
  --task "Audit Meta_Kim dependencies and update 4 READMEs" \
  --done "Audited npm dependency licenses" "Audited GitHub skill repositories" "Wrote English README dependency section" \
  --remaining "Update localized READMEs" "Push to GitHub" \
  --note "Found CLI-Anything is Apache 2.0, not MIT"

Output: show the user a concise progress summary such as "Progress saved: 2 completed, 2 remaining, 5 total sessions."

Layered Session Context (how it works)

The SessionStart hook outputs context in 3 layers to avoid context explosion:

LayerTriggerSizeContent
L1 compactAlways~120 charsTask state: "doing X, N done, M left"
L2 filteredProject tag matched~400 charsProject memories with relevance > 0.55
L3 fullUser queries manually~800 charsFull memories + recent

User can query on demand:

bash
# Query task state only (no MCP calls)
python mcp_memory_global.py --mode query-project

# Query all memories (full dump)
python mcp_memory_global.py --mode query-memories

If the user asks for previous context, history, or remembered notes, use --mode query-memories. If the user asks for project progress or where the previous session stopped, use --mode query-project."

Prompt Acceptance

This helper binds planning-continuity, memory-graph-and-observability, runtime-native-surfaces, and verification-eval-and-release. It saves resumable task context; it does not replace the canonical planning files or prove the user goal is done.

Pass

  • Saved state records the current task, completed work, remaining work, and any meaningful note.
  • The summary is specific enough for a later session to resume.
  • Existing task_plan.md, findings.md, and progress.md remain the active planning continuity source when present.

Fail

  • The command saves generic session noise or duplicates unchanged state.
  • It treats memory write success as task completion.
  • It overwrites planning files, PRD decisions, or canonical governance state.

Block

Block or report unavailable when the memory hook script cannot be found, Python is unavailable, the project directory is unknown, or the user asks for credential/private-state writes outside the approved memory boundary.

Return to stage

Return to Fetch when previous context is needed before saving. Return to Verification when the user asks whether saved state actually matches current project files.

Verification

Use the JSON output fields saved, file, total_sessions, last_session_completed, and last_session_remaining. For prompt acceptance, run npm run meta:prompt:validate.

Preserve

Preserve Claude Code native hooks, local memory, project planning files, MCP memory service state, filesystem boundaries, and user privacy. This command must not replace planning-with-files or Meta_Kim run artifacts.

Frequently asked questions

What does the Save Progress AI skill do?

Save current project task progress to local task state file for cross-session continuity. Use when the user asks to save progress, uses an equivalent localized trigger phrase, the session is interrupted, or the user wants to resume work later. Writes to .claude/project-task-state.json so next session can load it via SessionStart hook.

Why use Save Progress on TypingMind?

Because you install it once and use it with any model. Save Progress is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Save Progress in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/KimYx0207/Meta_Kim/tree/main/canonical/runtime-assets/claude/commands/save-progress. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Save Progress?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Save Progress?

As many as you like. As long as a model supports skills, you can use Save Progress with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Save Progress AI skill free?

Yes. It is published on GitHub by KimYx0207 under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇