a0p-fleet-runs
OfficialStandardize fleet runs with logs and SSE.
System Documentation
What problem does it solve?
Authoritative conventions for the a0p Fleet — how to start, log, observe, and terminate agent runs. It defines the contracts for the agent_runs and agent_logs schemas, the run_context ContextVars, the run_logger emit/flush pipeline, the SSE stream contract used by the Fleet UI, spawn caps, and the rules every new tool, orchestration mode, or sub-agent path must obey. Load this skill before adding any code that spawns a sub-agent, emits a structured log event, calls publish_log, or consumes /api/v1/runs.
Core Features & Use Cases
- Establishes the data models and schemas for agent_runs and agent_logs to enable robust run tracking and UI visualization.
- Defines the event vocabulary (spawn_start, spawn_complete, tool_call, tool_result, merge, cap_hit, provider_response, error, custom) and payload conventions to ensure consistent observability and debuggability.
- Describes the run_context, logging lifecycle, SSE contract, and cap rules to prevent phantom runs, leaks, and misattribution.
Quick Start
Load this skill before starting any sub-agent to enable correct run-spawn behavior, logging, and Fleet visibility.
Dependency Matrix
Required Modules
None requiredComponents
Standard package💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: a0p-fleet-runs Download link: https://github.com/The-Interdependency/a0/archive/main.zip#a0p-fleet-runs Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.