Discover AFS context roots, validate scaffolding, and read or write scratchpad memory.
npx skills add https://github.com/scawful/afs_scawful --skill agentic-context
Please help me install this Agent Skill. Skill: agentic-context Source: https://github.com/scawful/afs_scawful/tree/main/skills/agentic-context Command: npx skills add https://github.com/scawful/afs_scawful --skill agentic-context
This Skill helps automate safe AFS context handling by providing a structured approach to discover context roots, ensure necessary scaffolding, and perform read/write operations on the scratchpad while preserving memory integrity.
Locate the active context root and ensure scaffolding, then read or write to the scratchpad using the AFS CLI.
High-intent search queries and answers about installing and using this skill.
To discover the active context root, use standardized workspace navigation commands like ws list, ws find, and ws status. These tools locate the active project directory and validate the path before agents perform any memory read or write operations.
To safely write to scratchpad memory, use the AFS CLI wrapper to perform targeted write operations. The system enforces safe usage patterns by keeping all knowledge memory read-only except for authorized writes to the scratchpad area, preserving overall context integrity.
Context scaffolding is the required directory structure for secure cross-agent coordination. You need it because the system validates and ensures this scaffolding is present before use, allowing multiple agents to safely interact with the shared workspace without missing critical paths.
Yes, you can use workspace navigation tools such as ws list, ws find, and ws status for project discovery. These tools operate via an AFS CLI wrapper specifically designed to enable non-interactive agents to locate and navigate project contexts effectively.
The limitation of read-only memory is that agents cannot modify existing knowledge files directly. To persist new information, agents must route changes exclusively through the writable scratchpad, ensuring the broader context memory remains safe from unauthorized alterations or corruption.
zephyrwang6Analyze weekly L1 memory records to extract recurring patterns into L2 behaviors.
jholhewresManage persistent AI agents with shared memory, tasks, and notifications.
auxi-wardrobeDiscover files and gather codebase context using parallel search agents.
joemccannManage LLM context as persistent, governed artifacts in agentic AI projects.
