Atlas Framework
v1.0.0ATLAS Framework - Structured AI-assisted development methodology with GOTCHA 6-layer architecture and 5-step app building workflow. Use when building applica...
⭐ 0· 52·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name and description describe a development methodology; the SKILL.md contains guidance for architecture, workflow, memory, testing, and file layout. Requested actions (reading/writing memory files, initializing local SQLite DBs, validating integrations) align with a framework for building agentic apps.
Instruction Scope
The instructions explicitly direct the agent to read project-local files (memory/MEMORY.md, memory logs), check .env, validate API keys/OAuth flows, create folders, and initialize SQLite databases. This is coherent for a framework that maintains persistent memory and validates integrations, but it does mean the agent will be instructed to access and modify files in the project workspace. If you don't want the agent touching .env or other sensitive files, restrict those from the project context or sanitize them first.
Install Mechanism
No install spec and no code files are present (instruction-only). This minimizes risk from external downloads or arbitrary code installation.
Credentials
The skill does not declare required environment variables yet the instructions reference checking .env and verifying API keys/OAuth flows. For a framework that manages integrations this is reasonable, but you should confirm which credentials (if any) the agent will be given access to at runtime and avoid exposing unrelated or high-privilege secrets in project .env files.
Persistence & Privilege
always is false and the skill does not request platform-wide persistence or modify other skills. It does instruct creating project-local persistent memory and SQLite databases, which is expected for cross-session continuity and is scoped to the project.
Assessment
This skill is a methodology guide and is coherent with its stated purpose, but it instructs the agent to read and create project-local files (memory logs, .env) and to initialize SQLite DBs. Before installing or invoking it: (1) ensure your project .env does not contain unrelated/high-privilege secrets (move them out or use a dedicated credentials store), (2) run the skill in a sandboxed project directory if you want to limit file writes, (3) back up any existing memory/DB files you care about, and (4) confirm which integration credentials you will provide so the agent only verifies the specific services you intend. If you need more assurance, ask the skill author for an explicit list of exact files/paths it will read/write and the formats it expects.Like a lobster shell, security has layers — review code before you run it.
latestvk97e17fbvbmk9ms4t4wyg3m1js83qr1s
License
MIT-0
Free to use, modify, and redistribute. No attribution required.