Project Atlas AetherDev project hub
Atlas project concept

PrivateGPTConf

Package the steps and configuration needed to get a local PrivateGPT instance running with custom documents. Install prerequisites and clone PrivateGPT to a designated path. Configure Python environment (PyEnv) and dependencies (Poetry). Ingest a document folder and start the local service.

Type
Field Tool
Lifecycle
Prototyping
Last touched
2024-09-28
Visibility
Public

Purpose

Package the steps and configuration needed to get a local PrivateGPT instance running with custom documents.

Current state

Last touched: 2024-09-28. Functionality and completeness: Bootstrap scripts are present; documentation can be expanded for hardware/OS variations.

Next step

Add baseline automated tests to cover critical flows; Add CI pipeline for build/test/lint; Document deployment/runtime environment (or add Dockerfile); Document interfaces (CLI flags, API endpoints, file formats); Add structured logging and basic health checks.

Interfaces

Inputs
  • install path, document folder path, env/default template
  • Install paths
  • Document folders for ingestion
Outputs
  • Local PrivateGPT installation, ingested vector store

Reality to Action trace

Reality Ingestion

Contributes in this stage.

Canonical Storage

Contributes in this stage.

Automation Engines

Not in scope.

Human Interfaces

Contributes in this stage.

Operational Adoption

Contributes in this stage.

Core workflow

TBD. Document the 5-10 steps that define the core workflow.

Artifacts

  • N/A (bootstrap scripts)

Operational notes

Constraints and scars

  • Requires external downloads for models and dependencies; offline setup may need cached artifacts.

Reliability posture

Failure modes and safe behavior: Missing dependencies or network access prevent setup completion. Idempotency / retries / batching behavior: Re-running can re-clone or re-install; no built-in idempotency guards.

Observability

  • Logs: Script output plus PrivateGPT runtime logs
  • Metrics/health checks: None documented
  • Logs: script stdout/stderr and PrivateGPT runtime logs.

Security and privacy

Document folders can contain sensitive data; restrict access and avoid committing content.

Dependencies

Upstream
  • Optional model downloads from upstream sources

Ownership

Owners

Josh Barton

Users

Josh Barton (owner)

PrivateGPTConf

Architecture & Major Components

  • High-level diagram (text):

    • Entry/trigger -> core logic -> outputs (details per docs below)

  • Entry points: PrivateGPTSetup.sh, SetupDefault.sh

  • Top-level folders: data, env

  • Key abstractions: setup script, env/default template, ingestion path

Setup / Build / Run

  • Build system(s): Shell scripts and Python tooling.
  • Requires Git, Python 3.11, PyEnv, and Poetry; optional CUDA for GPU acceleration.