← Back to Skills Marketplace
AOI Triple Memory (Lite)
by
edmonddantesj
· GitHub ↗
· v0.1.2
891
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install aoi-triple-memory-lite
Description
AOI Triple Memory (Lite) — file search + decision notes templates (no plugins).
README (SKILL.md)
AOI Triple Memory (Lite)
S-DNA: AOI-2026-0215-SDNA-MEM01
What this is
A public-safe, plugin-free memory stack:
Provenance / originality
- AOI implementation is original code (no third-party code copied).
- Conceptually inspired by common “multi-layer memory” ideas (file notes + search + structured decisions).
- File-based memory:
MEMORY.md+memory/YYYY-MM-DD.md - Decision notes: structured
context/notes with tags - Fast search: ripgrep-based search across workspace
What this is NOT
- No external embeddings DB
- No automatic capture plugins
- No syncing to other machines
Commands
Search workspace
aoi-memory search --q "Tempo Hackathon" --n 20
Create a decision note (template)
aoi-memory new-note --title "Royalty rail decision" --tag royalty,base,usdc
Governance snippet (public)
We publish AOI skills for free and keep improving them. Every release must pass our Security Gate and include an auditable changelog. We do not ship updates that weaken security or licensing clarity. Repeated violations trigger progressive restrictions (warnings → publish pause → archive).
Support
- Issues / bugs / requests: https://github.com/edmonddantesj/aoi-skills/issues
- Please include the skill slug:
aoi-triple-memory-lite
License
MIT
Usage Guidance
This skill is largely what it claims to be (search + note templates), but check a few things before installing/using: 1) The code relies on the 'rg' (ripgrep) binary for searches, but the registry metadata says no binaries are required — ensure 'rg' is installed or searches will fail. 2) The search command reads all files under your workspace and prints matching lines/paths; do not run it in a directory containing secrets or sensitive configs unless you understand the output will include those matched lines. 3) The skill writes new notes into ./context in whatever workspace root is used (WORKSPACE env or current directory) — verify that location is acceptable. 4) No network access or credentials are requested, and there are no remote installs, so risk is limited to local file I/O; if you trust the code (you can review skill.js), it's probably safe to use in non-sensitive projects. If you want higher assurance, ask the author to update metadata to declare the 'rg' requirement or modify code to fallback to a pure-Node search if 'rg' is unavailable.
Capability Analysis
Type: OpenClaw Skill
Name: aoi-triple-memory-lite
Version: 0.1.2
The `skill.js` file contains a command option injection vulnerability in its `search` function. The user-provided query (`--q`) is passed directly to the `rg` (ripgrep) command via `node:child_process.spawnSync` without adequate sanitization. While `spawnSync` with an array of arguments generally prevents shell injection, a malicious query could still inject `rg`-specific options (e.g., `--files`, `--max-depth`, or potentially `--exec`) to alter the command's behavior or, in a worst-case scenario, lead to arbitrary command execution. There is no evidence of intentional malicious behavior like data exfiltration or backdoors, but this vulnerability poses a significant risk.
Capability Assessment
Purpose & Capability
Name/description, SKILL.md, and code all align on two features: workspace search and creating decision-note templates. However, the code calls the external 'rg' (ripgrep) binary for search while the registry metadata / requirements claim no required binaries. That is an internal inconsistency (the skill effectively requires 'rg').
Instruction Scope
SKILL.md and the binary behavior match: search runs ripgrep across the workspace and new-note writes a markdown file into ./context. The instructions do not reach outside the workspace or network. Note: search will read all files under the workspace and output matching lines — running this in a repo with secrets could expose sensitive content in the command output.
Install Mechanism
No install spec or remote downloads are present. The package is instruction-only with a local executable (skill.js) and no external dependencies — low install risk.
Credentials
The skill does not request credentials or environment variables, only optionally uses WORKSPACE (or cwd) as the root. This is proportionate, but because it reads files under the workspace and prints matches, users should be aware it can surface sensitive file contents if run in a sensitive directory.
Persistence & Privilege
The skill is not always-enabled and does not modify other skills or system-wide settings. It writes files only under the workspace 'context/' directory (expected for a note-taking tool).
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install aoi-triple-memory-lite - After installation, invoke the skill by name or use
/aoi-triple-memory-lite - Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.2
Docs: add GitHub Issues support link (aoi-skills).
v0.1.1
Docs: add provenance/originality statement (no third-party code copied).
v0.1.0
Initial plugin-free memory stack: rg search + decision note template. MIT.
Metadata
Frequently Asked Questions
What is AOI Triple Memory (Lite)?
AOI Triple Memory (Lite) — file search + decision notes templates (no plugins). It is an AI Agent Skill for Claude Code / OpenClaw, with 891 downloads so far.
How do I install AOI Triple Memory (Lite)?
Run "/install aoi-triple-memory-lite" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is AOI Triple Memory (Lite) free?
Yes, AOI Triple Memory (Lite) is completely free (open-source). You can download, install and use it at no cost.
Which platforms does AOI Triple Memory (Lite) support?
AOI Triple Memory (Lite) is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created AOI Triple Memory (Lite)?
It is built and maintained by edmonddantesj (@edmonddantesj); the current version is v0.1.2.
More Skills