Skip to main content

Agent Resurrect (agent-resurrect-btw)

PyPI Track

Agent Resurrect is an autonomous telemetry and circuit-breaking engine for AI coding agents (Claude Code, Codex, Gemini CLI). It prevents infinite repair "doom-loops", safeguards API budgets, and retrieves team-wide resolutions stored in Databricks Delta Lake.


Quick Start

1. Installation

Install the SDK from PyPI:

pip install --upgrade agent-resurrect-btw

2. Initialize in Your Repository

Run in your project root:

resurrect init

This enables Entire checkpoint tracking and creates a starter configuration file at:

.resurrect/config.json

3. Configure Databricks Credentials

Open .resurrect/config.json and fill in your connection details:

{
  "databricks_host": "https://<your-workspace>.cloud.databricks.com",
  "databricks_token": "dapi...",
  "databricks_warehouse_id": "<your-warehouse-id>",
  "delta_table_name": "dev_failure_traps"
}

Alternative: Environment Variables

You can also supply credentials via your environment without editing files:

export DATABRICKS_HOST="https://<your-workspace>.cloud.databricks.com"
export DATABRICKS_TOKEN="dapi..."
export DATABRICKS_WAREHOUSE_ID="<your-warehouse-id>"
export DELTA_TABLE_NAME="dev_failure_traps"

4. Run Pre-Flight Diagnostics

Validate connectivity to Git, Entire CLI, and Databricks SQL Serverless:

resurrect doctor

For deep-dive logs and command traces, append --verbose:

resurrect doctor --verbose

Usage

Live Agent Execution & Circuit Breaking

Execute your coding agent or test suite wrapped in the Resurrect telemetry harness:

resurrect run "pytest test_redis_worker.py"

Or with an interactive agent:

resurrect run "codex 'fix the failing cluster connection in worker.py'"

How It Works:

  • Live Subprocess Supervision: Resurrect streams agent output live in the terminal.
  • Loop Detection: Evaluates consecutive failure signatures from Entire Git checkpoints (refs/entire/checkpoints/).
  • Emergency Circuit Breaker: When failure counts reach 3, execution halts immediately to prevent token and credit burn.
  • Lakehouse Remediation: Fetches proven fixes from Databricks Delta Lake (dev_failure_traps).
  • Handoff Manifest: Generates .resurrect/handoff_manifest.json with token counts, cost saved estimates, and context badges.

Inspect Checkpoints & Telemetry

Inspect the latest session checkpoint, token estimates, and privacy status:

resurrect inspect-checkpoint

Track 1: Privacy Boundary Compliance

Agent Resurrect includes a local security and sanitization boundary:

  • Zero Leakage: Raw prompts, full stack frames, and transcripts are never sent to external services.
  • Token & Path Scrubbing: Local paths (C:\Users\...) and secrets (dapi..., Bearer...) are scrubbed before SQL query generation.
  • Explicit Context Transparency: Output is badged as [COMPLETE CONTEXT] or [PARTIAL / REDACTED CONTEXT] so downstream engineers know whether fixes are verified or heuristic.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

agent_resurrect_btw-0.1.3.tar.gz (15.8 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

agent_resurrect_btw-0.1.3-py3-none-any.whl (17.2 kB view details)

Uploaded Python 3

File details

Details for the file agent_resurrect_btw-0.1.3.tar.gz.

File metadata

  • Download URL: agent_resurrect_btw-0.1.3.tar.gz
  • Upload date:
  • Size: 15.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/7.0.0 CPython/3.11.0

File hashes

Hashes for agent_resurrect_btw-0.1.3.tar.gz
Algorithm Hash digest
SHA256 ec95b0fef8a45cf1f597f6a707b76e68a06fd4adc281752a5d9e70941d955ee2
MD5 a8d0b873d9269b6017f1c39b014ee731
BLAKE2b-256 1fcccd8b8cf18ce26e0aa9c89bb34627968832261faed12f9d183e3db359d3a9

See more details on using hashes here.

File details

Details for the file agent_resurrect_btw-0.1.3-py3-none-any.whl.

File metadata

File hashes

Hashes for agent_resurrect_btw-0.1.3-py3-none-any.whl
Algorithm Hash digest
SHA256 f5b24d1436cac0c3e41ae282caab3e970b446d96f4097f271951adc4f0ffffa7
MD5 1017b41755d8bfc54ac1a990fcb8d821
BLAKE2b-256 f0e50322b8c2e75b44700cfe8fb454f3b9121c3177d4e9fbe25c4a09c6d66c90

See more details on using hashes here.

Release history Release notifications | RSS feed

This release

0.1.3 This release

2 files

0.1.2

2 files

0.1.1

2 files

0.1.0

2 files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page