Memory

A lightweight, open-source AI assistant built for local hardware.

Install

npx -y skills add cutec-chris/PawLia --skill memory

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Long-term memory — search everything from past conversations. ALWAYS use this skill when the user: asks what was said before, references earlier conversations, wants to know or remember something from the past, asks "what did I/we/you say about...", "do you remember...", "have we talked about...", "what was that thing about...", or any question that requires knowledge from previous days/sessions. Also use when YOU are unsure whether a topic was discussed before. Commands: "index" to reindex, "status" to check index state, "dream" to trigger wiki consolidation, "lint" for wiki health check.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

3.8 KB, as published. Nobody here has run it

Memory Skill

How to use

The query is either a search question or a management command. Run via the Bash tool. The user ID is automatically provided via the PAWLIA_USER_ID environment variable — do NOT pass it manually.

python <scripts_dir>/memory.py search "<question>"
python <scripts_dir>/memory.py index
python <scripts_dir>/memory.py dream
python <scripts_dir>/memory.py lint
python <scripts_dir>/memory.py status

Commands

Command	Bash call	Description
search	`python <scripts_dir>/memory.py search "<question>"`	Semantic search across all past conversations
index	`python <scripts_dir>/memory.py index`	Index any new/updated daily chat logs
dream	`python <scripts_dir>/memory.py dream`	Trigger Dream Wiki: process chat logs into structured wiki pages
lint	`python <scripts_dir>/memory.py lint`	Wiki health check: merge overlapping pages, fix missing links
status	`python <scripts_dir>/memory.py status`	Show how many days are indexed

Step-by-step instructions

Determine if the query is a command (index, status, dream, lint) or a search question.
For search questions, use the search command with the user's question.
Run the command using the Bash tool.
Return the result — for searches, present the relevant information naturally.

Important

Every search keyword-scans the full daily-log history straight from the chat logs (no index needed), so any past conversation is recallable — today's, a call/file from minutes ago, and threads from weeks back that were never distilled into the wiki. Only threads that actually share keywords with the question are returned; if nothing matches you get a clear "nothing found" instead of unrelated context.
On top of that, search queries the distilled long-term wiki for consolidated knowledge. Background indexing/dreaming runs automatically on idle — you do not need to index before a search.
Search results contain relevant excerpts from past conversations — use them to answer the user's question.
The dream command processes unprocessed chat logs into a structured wiki with cross-references (Obsidian wikilinks by default, configurable via wiki_link_format).
The Dream Wiki runs automatically when the user has been idle for 30 minutes (configurable via dream_idle_minutes).