No red flags found in any of the 11 categories — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →
by mrvladd-d · Codex Skill · ★ 54
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
🔒 Is memobank safe to install? View the security audit →
memobank Language: English | Русский is a source-only skill pack and workflow framework for AI-first software development in Codex CLI, Claude Code, OpenCode, and compatible agent runtimes. It keeps project context in repository files instead of fragile chat history. An agent can read those files, understand the current project state, continue a task in a later session, and leave evidence for the next run. When to use it Use when: a project will take more than one short chat session; requirements, architecture, or task status must survive context resets; several agents or fresh sessions need the same source of truth; you want PRD, specifications, tasks, implementation, and verification to stay connected; you want manual control first, then optional automation later. Do not start with automation if the product idea is still unclear. Start with the manual workflow and let the files become the shared project memory.
| Stars | 54 |
| Forks | 11 |
| Language | JavaScript |
| Category | Codex Skill |
| License | MIT |
| Quality Score | 64.3345722823544/100 |
| Last Updated | 2026-06-02 |
| Created | 2026-03-05 |
| Platforms | claude-code, cli, codex, node |
| Est. Tokens | ~168k |
Explore other popular codex skill tools:
memobank is Agent-first Memory Bank skill pack for Codex CLI, Claude Code, and similar runtimes.. It is categorized as a Codex Skill with 54 GitHub stars.
memobank is primarily written in JavaScript.
You can find installation instructions and usage details in the memobank GitHub repository at github.com/mrvladd-d/memobank. The project has 54 stars and 11 forks, indicating an active community.
memobank is released under the MIT license, making it free to use and modify according to the license terms.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: