Skip to main content

A graph-based task management framework for AI agents

Project description

🌌 SocialSeed Tasker

The Operational Brain for AI-Driven Engineering

SocialSeed Tasker is the definitive framework for teams and organizations scaling their development with AI Agents. While agents provide speed, Tasker provides the Architectural Guardrails that prevent your codebase from descending into chaos.

Version Architecture Storage


🚀 Why SocialSeed Tasker?

In the age of autonomous coding, speed is no longer the bottleneck—context and safety are. Tasker turns your repository into a living, reasoning graph that AI agents can navigate with precision, ensuring every change is calculated, documented, and safe.

🛡️ Develop with Agents, Without the Breakage

Agents can't see what they might break three layers deep. Tasker can. Our Graph-Based Impact Analysis maps every dependency, blocking risky changes before they happen.

🧠 Infinite Context for LLMs

Stop feeding agents raw text. Tasker's Code-as-Graph (powered by Tree-Sitter) provides agents with a semantic map of your functions, classes, and imports, reducing "hallucinations" and increasing success rates.

👁️ Human-in-the-Loop Visibility

With AI Reasoning Logs, you see exactly why an agent made a decision. No more black boxes—just clear, traceable engineering logic.


💎 Key Features for the Modern Tech Stack

  • Graph-Native Management: Powered by Neo4j, managing complex microservices dependencies is as simple as a single query.
  • Impact Analysis Engine: Instantly calculate the "Blast Radius" of any code change or issue resolution.
  • RAG-Powered Discovery: Semantic search across your entire task history and codebase to find similar solutions in seconds.
  • Architectural Constraints: Set "Hard Rules" (e.g., "No circular dependencies", "Must use PostgreSQL") that both agents and humans must follow.
  • Vue-Powered Observability: A stunning Kanban and Graph dashboard to visualize your project's health in real-time.

🤖 Quick Start for AI Agents

Getting your agents up and running with Tasker takes seconds.

# 1. Install the Brain
pip install socialseed-tasker

# 2. Spin up the Infrastructure
docker compose up -d

# 3. Give the Agent Context (The Magic Part)
tasker agent context --issue "ISSUE_ID"

This single command provides an agent with the specific code symbols, dependencies, and historical context needed to solve the task safely.


🏢 Built for Scale

Whether you are a solo developer managing a dozen agents or an enterprise architect overseeing 50 microservices, Tasker scales with you.

  • CTOs & Tech Leads: Enforce architectural standards automatically across all teams.
  • DevOps: Monitor system stability and root-cause failures through causal graph inference.
  • Product Managers: Get real-time, data-driven insights into project blockers and critical paths.

📚 Full Documentation

For detailed technical guides, API references, and architectural deep-dives, visit our official documentation site:

👉 https://daironpf.github.io/socialseed-tasker/

📖 v0.9.0 Documentation


🔧 Troubleshooting

Neo4j Container Fails to Start

If you see the error: Invalid value for password. It cannot be 'neo4j', which is the default.

Cause: Neo4j 5.x security policy prevents using "neo4j" as the password (it must differ from the username).

Solution: Use a different password in docker-compose.yml:

environment:
  - NEO4J_AUTH=neo4j/neoSocial  # password must NOT be "neo4j"

Also ensure the API environment variables match:

- TASKER_NEO4J_PASSWORD=neoSocial

Restart after changing:

docker compose down -v
docker compose up -d

🤝 Contributing

We are building the future of autonomous engineering. Join us!


"Speed is a byproduct of clarity. Tasker provides the clarity."

Project details


Download files

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

Source Distribution

socialseed_tasker-0.9.1.tar.gz (380.6 kB view details)

Uploaded Source

Built Distribution

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

socialseed_tasker-0.9.1-py3-none-any.whl (412.8 kB view details)

Uploaded Python 3

File details

Details for the file socialseed_tasker-0.9.1.tar.gz.

File metadata

  • Download URL: socialseed_tasker-0.9.1.tar.gz
  • Upload date:
  • Size: 380.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for socialseed_tasker-0.9.1.tar.gz
Algorithm Hash digest
SHA256 e407242a19f483222b26cfb074c6c41b8b6c761fd8ffb5eb099dca72951ebb99
MD5 bef6f85db7bb5e784bfebfe97305cff2
BLAKE2b-256 367f3f6fa86e4de53a3ca7ff3ea0d0bd71c5c85aa0ae84359e0bb13d8768a71e

See more details on using hashes here.

File details

Details for the file socialseed_tasker-0.9.1-py3-none-any.whl.

File metadata

File hashes

Hashes for socialseed_tasker-0.9.1-py3-none-any.whl
Algorithm Hash digest
SHA256 291b218e27044653db7a8cc1328436b71a6b0c0ce9d5c2f2b73ff376274d1a58
MD5 7b2baa13399254211f077e899cb1b405
BLAKE2b-256 631b007c6cedbc43b61c295d9561041b4052d7cfabd499d1f1adf040b73fcc1f

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page