feat(templates): add conversation history persistence to HTTP agent templates

[aidandaly24]

Description

All four HTTP agent templates (Strands, OpenAI Agents, Google ADK, LangChain/LangGraph) were stateless per-invocation — each call to the agent started with zero conversation context. This meant users couldn't have multi-turn conversations where the agent remembers what was said in previous turns within the same session.

This PR adds conversation history persistence using each framework's built-in session management mechanism:

• Strands: Per-session Agent cache keyed by context.session_id. Strands Agent accumulates messages internally in self.messages across stream_async() calls, so reusing the same instance per session gives natural multi-turn memory. This also fixes a conversation bleed bug where the previous single global _agent shared history across all sessions/users.
• OpenAI Agents: SQLiteSession (from the SDK) passed via the session= parameter on Runner.run(). The session automatically loads prior history before inference and saves new turns after. Uses in-memory SQLite by default (:memory:).
• Google ADK: Module-level InMemorySessionService with a get_or_create_session() helper that calls get_session() first and only create_session() if the session doesn't exist (avoids AlreadyExistsError on repeat calls). The module-level Runner shares the same service instance across invocations.
• LangGraph: Module-level InMemorySaver checkpointer with thread_id mapped to context.session_id. The add_messages reducer on AgentState.messages automatically appends new HumanMessages to the existing conversation from the checkpoint. The graph is recreated per-invocation (to pick up dynamic MCP tools) but the checkpointer is shared, so state persists.

Also adds missing system prompts:

• OpenAI Agents: Added instructions="You are a helpful assistant. Use tools when appropriate." to all Agent() constructors (previously had no system prompt at all).
• LangGraph: Added prompt="You are a helpful assistant. Use tools when appropriate." to create_react_agent().

All conversation state is in-memory (best-effort) — it persists across invocations within the same runtime process but resets on cold starts. This is the correct default for starter templates; users can upgrade to durable backends (Strands FileSessionManager, OAI SQLiteSession with file path, ADK DatabaseSessionService, LangGraph persistent checkpointers) as needed.

Related Issue

N/A — discovered during session storage bugbash testing

Type of Change

• Bug fix
• New feature
• Breaking change
• Documentation update
• Other (please describe):

Testing

How have you tested the change?

• I ran npm run test:unit and npm run test:integ
• I ran npm run typecheck
• I ran npm run lint
• If I modified src/assets/, I ran npm run test:update-snapshots and committed the updated snapshots

End-to-end deploy + invoke testing:

• Strands (Bedrock): Deployed, invoked 3x — recall ✅, session isolation ✅
• OpenAI Agents (OpenAI): Deployed, invoked 3x — recall ✅, session isolation ✅
• LangGraph (Bedrock): Deployed, invoked 3x — recall ✅, session isolation ✅
• Google ADK (Gemini): Deployed, code structurally verified — Gemini API quota prevented live invocation testing

Checklist

• I have read the CONTRIBUTING document
• I have added any necessary tests that prove my fix is effective or my feature works
• I have updated the documentation accordingly
• I have added an appropriate example to the documentation to outline the feature, or no new docs are needed
• My changes generate no new warnings
• Any dependent changes have been merged and published

---

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the
terms of your choice.

[github-actions]

Coverage Report

Status	Category	Percentage	Covered / Total
🔵	Lines	44.35%	6987 / 15753
🔵	Statements	43.82%	7415 / 16919
🔵	Functions	41.82%	1251 / 2991
🔵	Branches	43.26%	4687 / 10832

Generated in workflow #68 for commit b6dd2a6 by the Vitest Coverage Report Action