Procedural wiki
Rules, identities, hard boundaries, project-wide truths, and standards that should never drift.
Sovereign agent memory
Vraksha is a local-first agent that restores your rules, project state, and personal context before every session. No amnesia. No borrowed memory. Just continuity on your machine.
$ curl -fsSL https://raw.githubusercontent.com/vraksha/vraksha/main/install-linux.sh \
| bash
The thing most agents fake
Vraksha keeps durable knowledge in a local tri-store foundation so every conversation begins with your actual world loaded.
Rules, identities, hard boundaries, project-wide truths, and standards that should never drift.
Past decisions and session context are indexed for retrieval instead of left behind in dead chats.
Entities, projects, trust hints, and references prepare the path toward graph-backed reasoning.
Guarded autonomy
Vraksha is being built for machine operations with real scrutiny: sanitize inputs, verify intent, score actions, and keep execution isolated.
Prompt injection and untrusted instructions are screened before they touch the reasoning loop.
Shell commands, file writes, and external operations move through trust-scored checks.
Docker isolation is the baseline today, with stronger sandbox targets on the roadmap.
Already useful
Memory consolidation and logging stay away from the reasoning hot path.
New experts load from folders with skill.py and SKILL.md.
A specialized analyzer helps detect AI-generated code artifacts during review.
Use hosted frontier models, OpenRouter, Bedrock, Groq, Ollama, and local endpoints.
Where it is going