DataMind v0.3.0 Release

DataMind v0.3.0

DataMind v0.3.0 is a preview release focused on safer local execution,
scope-aware memory, and a complete PyPI installation path.

Preview release: the core architecture is functional, but some
integrations are still being expanded. For production use, start with an
isolated profile and validate the target database, model gateway, and
deployment environment first.

Installation

pip install datamind

Optional extras:

pip install 'datamind[mysql]'         # MySQL dialect support
pip install 'datamind[postgres]'      # PostgreSQL dialect support
pip install 'datamind[voyage]'        # Voyage embeddings
pip install 'datamind[huggingface]'   # Local BGE / e5 embeddings

Highlights

Scope-Typed Memory v2

Memory entries now carry explicit scope, kind, and status fields:

• scope: global, profile, or session
• kind: preference, decision, workflow, summary, skill, or fact
• status: active or archived

Recall combines scope-conditioned retrieval results, which helps keep
profile-specific preferences from leaking across tenants or sessions. Existing
v0.2 memory stores are migrated automatically on first open.

Hook-Based Safety Layer

Tool execution now runs through a HookChain that can return Allow, Deny,
AskUser, or Rewrite. Three hooks are enabled by default:

• PathAllowlistHook resolves symlinks and blocks file access outside the
  active working directory or profile data directory.
• DestructiveSqlHook uses sqlglot to detect destructive SQL operations.
  DELETE, UPDATE, DROP TABLE, TRUNCATE, and ALTER require
  confirmation; DROP DATABASE and DROP SCHEMA are denied.
• AuditLogHook records tool calls in storage/<profile>/audit.jsonl with
  a hash chain for tamper detection. Secret-shaped fields such as api_key,
  password, and token are redacted automatically.

The browser UI now displays a confirmation modal when user approval is needed.
Streaming responses also include hook_asks_user and hook_denied events.

PyPI Packaging

The PyPI package now includes the runtime code, browser UI, and default skill
catalog needed for a standard pip install datamind workflow:

• datamind/ Python package
• datamind/static/app.html
• datamind/skills/{code-review,db-ops-sop}/SKILL.md

Release publishing is configured through GitHub Actions with PyPI Trusted
Publishing, so package uploads do not require long-lived PyPI tokens.

Fixes

• Fixed an empty-wheel packaging issue caused by an overly broad package
  exclusion rule.
• Switched packaging metadata to source the version from datamind.__version__.
• Aligned project metadata with the repository license (Apache-2.0).

Quick Start

pip install datamind

export DATAMIND__LLM__API_BASE=https://your-gateway.example.com
export DATAMIND__LLM__API_KEY=sk-...
export DATAMIND__LLM__MODEL=claude-sonnet-4-6

datamind chat

To run the browser UI locally:

python -m uvicorn datamind.server:app --port 8000
# Open http://127.0.0.1:8000

Verification

Automated test suite:

133 passed in <1s

Smoke checks:

python -m datamind.scripts.hello_sdk      # model gateway connectivity
python -m datamind.scripts.hello_memory   # scope-typed memory and extraction
python -m datamind.scripts.hello_hooks    # sandboxing and audit hash chain

Known Limitations

• BYOP HTTP sinks: chunk, vector, and graph payload schemas are defined, but
  HTTP sink endpoints and adapter examples are still in progress. Current
  ingest flows accept local file paths.
• MySQL / PostgreSQL: dialect support is available through optional extras
  and smoke checks, but users should run environment-specific validation before
  production use.
• SDK backend: DATAMIND__AGENT__BACKEND=sdk is available, but hook-chain
  enforcement currently runs through the native execution loop.
• Large deployments: this preview has not yet been validated against
  large-scale production workloads.

Roadmap

• HTTP sink endpoints for chunk, vector, and graph ingestion.
• Additional ETL adapter examples for common orchestration tools.
• Broader database validation for MySQL and PostgreSQL.
• Expanded safety and recovery evaluations for hook-governed execution.

Links

• PyPI: https://pypi.org/project/datamind/
• Issues: https://github.com/OpenDCAI/DataMind/issues

Bug reports are welcome during the preview. Please include the relevant smoke
check output and a redacted Settings() snapshot when possible.