Coordinate planning, coding, review, QA, merge, and verification agents across repositories.
npx skills add https://github.com/dawsonblock/AAE_ORACLE --skill swe-af
Please help me install this Agent Skill. Skill: swe-af Source: https://github.com/dawsonblock/AAE_ORACLE/tree/main/aae-engine/vendor/SWE-AF-main/docs Command: npx skills add https://github.com/dawsonblock/AAE_ORACLE --skill swe-af
SWE-AF solves the challenge of coordinating a team of AI agents to scope, plan, build, review, and verify software across repositories, enabling end-to-end automated software engineering at scale.
Trigger a SWE-AF build by posting an execute request to the API with the target repo and runtime configuration.
High-intent search queries and answers about installing and using this skill.
Multi-agent orchestration coordinates planning, coding, QA, merging, and verification agents across monorepos and multi-repo setups using a shared memory for knowledge propagation and a single API call. It manages parallel execution and sequential dependent tasks automatically.
DAG-based parallel execution runs independent tasks simultaneously at specific levels with deterministic gating and barrier synchronization. Dependent tasks execute sequentially, ensuring efficient automated software engineering workflows while maintaining strict execution order.
Runtime plan mutation allows modifying execution plans dynamically during automated software engineering. It includes debt tracking to gracefully degrade work and propagates context across agents, ensuring builds adapt to changing requirements without restarting.
AI coding agents handle worktree isolation by separating execution environments for parallel tasks. Durable checkpoints save progress at each DAG stage, enabling recovery and verification across planning, coding, review, and merge phases without data loss.
Multi-agent software orchestration supports governance and audit trails via AgentField integration. It provides auditable provenance for all automated actions across cross-repo builds, ensuring accountability for planning, coding, and verification stages.
Parallel DAG execution limitations for multi-repo AI builds include barrier synchronization overhead and strict dependency sequencing. While independent levels run in parallel, dependent tasks must wait, and runtime plan mutation may introduce debt that requires manual resolution.
Agent-FieldCoordinate autonomous AI agents to plan, code, and verify software builds.
chuangkevinExecute written implementation plans with verification checkpoints and task state transitions.
FFCfelps1Plan, execute, and ship software features with code review and CI/CD integration.
wtyler2505Coordinate Claude Code agents across isolated worktrees with dependency DAG enforcement.
