Decouple AI agent runtime construction

[luisorofino]

What does this PR do?

Refactors the AI framework runtime construction path so callers pass a single AgentRuntime through the ReAct flow instead of manually carrying separate agent and tool registry objects.

This PR:

• Replaces worker/subagent/goal-specific builders in ddev.ai.agent.build with one generic AgentRuntimeFactory.build_runtime(...) API.
• Keeps AgentRuntime as the pair of agent + tool registry, and updates ReActProcess to accept that runtime directly.
• Moves orchestration-specific policy out of the agent layer:
  • AgenticPhase renders the worker system prompt and provides the selected parent AgentConfig.
  • phases.goal derives the reviewer config, including read-only tools and the reviewer prompt.
  • SpawnSubagentTool derives child agent configs and allowed child tools from the parent config.
• Simplifies ToolRegistry context by removing subagent-specific builder/allowed-tool plumbing and passing generic runtime infrastructure to tool factories.
• Updates AI framework tests for runtime construction, ReAct process wiring, goal validation, subagent spawning, and registry construction.

Motivation

The previous design introduced AgentRuntime, but ReActProcess and several call sites still unpacked runtime objects into separate agent and tool-registry values. The agent build layer also knew too much about goal reviewers and subagents, which made phase policy leak into generic agent construction.

This refactor keeps agent.build focused on constructing runtimes from explicit inputs, while goal and subagent-specific behavior stays in the modules that own those workflows. The result is a smaller builder API, clearer ownership boundaries, and less duplicated runtime wiring.

Review checklist (to be filled by reviewers)

• Feature or bugfix MUST have appropriate tests (unit, integration, e2e)
• Add qa/required if this PR needs QA validation, or qa/skip-qa if it does not. Exactly one of the two is required.
• If you need to backport this PR to another branch, you can add the backport/<branch-name> label to the PR and it will automatically open a backport PR once this one is merged

[datadog-official]

   

✨ Fix all issues with BitsAI

⚠️ Warnings

🚦 5 Pipeline jobs failed

PR All | test / j06ca546 / SNMP     

PR All | test / j91231ff (py3.13-6.0-standalone) / MongoDB-py3.13-6.0-standalone     

PR All | test / j9eb6aa6 / DuckDB     

View all 5 failed jobs.

ℹ️ Info

No other issues found (see more)

🧪 All tests passed
❄️ No new flaky tests detected

🎯 Code Coverage (details)
• Patch Coverage: 98.86%
• Overall Coverage: 87.91%

Useful? React with 👍 / 👎

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 945cd70 | Docs | Datadog PR Page | Give us feedback!}

[luisorofino]

• Add max_timeout to PhaseOrchestrator's arguments #24010
• Add context flags to TaskConfig #23992
• Centralize agent logging behind scoped callbacks and remove artifact_root tramp data #23935
• Improve readability of run_tasks in AgenticPhase #23919
• Move AI orchestration runtime out of phases #23916
• Decouple AI agent runtime construction #23912 👈 (View in Graphite)
• Replace extra_init_kwargs with polymorphic Phase.build() factory and ResourceProvider #23889
• loa/openmetrics-ai-gen

This stack of pull requests is managed by Graphite. Learn more about stacking.

[dd-octo-sts]

Validation Report

All 21 validations passed.

Show details

Validation	Description	Status
agent-reqs	Verify check versions match the Agent requirements file	✅
ci	Validate CI configuration and code coverage settings	✅
codeowners	Validate every integration has a CODEOWNERS entry	✅
config	Validate default configuration files against spec.yaml	✅
dep	Verify dependency pins are consistent and Agent-compatible	✅
http	Validate integrations use the HTTP wrapper correctly	✅
imports	Validate check imports do not use deprecated modules	✅
integration-style	Validate check code style conventions	✅
jmx-metrics	Validate JMX metrics definition files and config	✅
labeler	Validate PR labeler config matches integration directories	✅
legacy-signature	Validate no integration uses the legacy Agent check signature	✅
license-headers	Validate Python files have proper license headers	✅
licenses	Validate third-party license attribution list	✅
metadata	Validate metadata.csv metric definitions	✅
models	Validate configuration data models match spec.yaml	✅
openmetrics	Validate OpenMetrics integrations disable the metric limit	✅
package	Validate Python package metadata and naming	✅
qa-label	Validate the pull request declares whether it needs QA for the next Agent release	✅
readmes	Validate README files have required sections	✅
saved-views	Validate saved view JSON file structure and fields	✅
version	Validate version consistency between package and changelog	✅

View full run