Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

feat(starknet): session-key spending policies for AI agents

Open
  • Is something wrong?
#31 0 comments 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
5/5
Estimated time
Over a week
Newbie friendliness
25/100
Issue type
Feature
Clarity
Mostly clear
Activity status
Active
Tech stack
python
Domain
blockchain

Research direction

Start by reading issue #2 and the agent configuration and funding paths named here: ai_agent_service._execute_funding and agent/cost_guard.py. Verify the current session-key standard, maintained Python client API, and testnet paymaster availability before choosing an account implementation. Done means policy limits are enforced on-chain, session keys can be revoked and expire, and policy parameters map to existing agent config fields, with tests and documentation.

Written by the indexing model from the issue text.

Description

agents chain-integration enhancement tier-3

Summary

Enforce AI agent spending limits in a Starknet smart account via session keys, so caps hold even if the backend is compromised, and offer gasless onboarding through a paymaster.

Motivation

Agent budgets (max_per_proposal, daily_budget) are enforced only in application code. On-chain policy is a stronger guarantee and a good showcase of account abstraction.

Current state in this repo

  • Limits live in agent config documents and ai_agent_service._execute_funding; agent/cost_guard.py enforces a separate set of budgets for the workspace agent.

Proposal

  • Register Starknet testnet in the registry (#2) with a minimal adapter.
  • Select an account implementation supporting session keys and define a policy (allowed targets/tokens, per-call and per-day limits, expiry).
  • Create the session key for an agent and use it for agent-initiated funding calls; revoke on demand from the UI.
  • Optionally sponsor fees with a paymaster for new users.
  • Document the toolchain (Cairo, Python client) and its cost for contributors.

Acceptance criteria

  • An agent can transact within policy and is rejected on-chain beyond it.
  • Session keys can be revoked and expire automatically.
  • Policy parameters map to the existing agent config fields.

Out of scope

  • Proving the quantum simulation with STARKs (research-grade).
  • Mainnet.

Technical notes and risks

  • Session-key standards are still a draft and wallet implementations are not mutually compatible. Pick one implementation and document the lock-in.

Dependencies

  • #2

Verify before implementing

  • Current session-key standard status and the maintained Python client API.
  • Paymaster availability on testnet.

Part of ##1. Ground rules for all issues in this roadmap are listed there (testnet-first, no secrets in the repo, chain logic behind the registry, tests and docs with every PR).

Dominant language
Python
Stars
2
Forks
4
Avg merge
6h 54m
Merged PRs (30d)
7

Getting set up

  • Ships a Dockerfile or Docker Compose file
  • No pull request template
  • No contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from QMeshPy/distributed-quantum

All issues in QMeshPy/distributed-quantum

Similar issues

More Python issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.