Issue #974: Remove dead code in langchain_agent (summarize_conversation + helpers.py) (#975)

##### Description of Change

Pure dead-code removal in the `langchain_agent` component, found during
a code review.

- **`summarize_conversation()` in `core.py`** referenced a non-existent
`self.agent` attribute (the class stores agents in `self.agents`, a
dict). It would have raised `AttributeError` if ever called — but it has
**no callers** anywhere in `components/` or `bases/`. Removed.
- **`helpers.py` (entire module)** — `update_person_interactions()` and
its private `_build_mutation_query()` were **unreferenced** across the
repo (nothing imports `helpers`). Interaction summaries are persisted
via the `lif_mutation` MCP tool driven by the `save_interaction_summary`
prompt, not this helper. Removed the file.
- Updated the component README layout table to drop the `helpers.py`
row.

No behavior change for any current caller, since neither symbol was
referenced. The public surface (`LIFAIAgent`) is unchanged.

**How to test:** `uv run pre-commit run --all-files` — ruff, cspell, ty,
and the full pytest suite pass. No runtime path is affected.

##### Related Issues

Closes #974

##### Type of Change

- [x] Code refactoring

##### Project Area(s) Affected

- [x] components/

---

##### Checklist

- [x] commit message follows commit guidelines (see
commitlint.config.mjs)
- [x] code passes linting checks (`uv run ruff check`)
- [x] code passes formatting checks (`uv run ruff format`)
- [x] code passes type checking (`uv run ty check`)
- [x] pre-commit hooks have been run successfully

##### Testing

- [x] Automated tests added/updated (existing suite green; no new
behavior to test)

##### Additional Notes

Dead code only — no tests added since there is no new behavior, and the
removed code had no test coverage or callers.

---------

Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>

Author: bjagg

components/lif/langchain_agent/README.md

@@ -7,7 +7,6 @@ Wraps LangChain/LangGraph into a `LIFAIAgent` purpose-built for the Advisor's ch
 | File | Contents |
 |---|---|
 | `core.py` | `LIFAIAgent` — top-level interface (`setup`, `ask_agent`) |
-| `helpers.py` | Prompt + chain construction helpers |
 | `memory.py` | LangGraph memory wiring (`langmem`-backed) |
 | [`prompts/`](prompts/) | Text-file prompt templates loaded at runtime |
 

components/lif/langchain_agent/core.py

@@ -10,7 +10,6 @@
 
 Environment Variables:
     - TOOL_URL: URL for MCP tools
-    - GRAPHQL_URL: URL for GraphQL endpoint
     - LLM_MODEL_NAME: Name of the LLM model to use
 """
 
@@ -29,7 +28,6 @@
 
 
 LIF_SEMANTIC_SEARCH_MCP_SERVER_URL = os.environ.get("LIF_SEMANTIC_SEARCH_MCP_SERVER_URL")
-LIF_GRAPHQL_API_URL = os.environ.get("LIF_GRAPHQL_API_URL")
 LLM_MODEL_NAME = os.environ.get("LIF_ADVISOR_LLM_MODEL_NAME")
 LLM_TOKEN_COSTS = {
     "gpt-4o-mini": {"input": 1.1, "output": 4.4, "cached": 0.275},
@@ -207,32 +205,6 @@ async def ask_agent(self, task, message: str) -> dict:
         logger.info(f"Response: {response}")
         return response
 
-    async def summarize_conversation(self, config: dict) -> dict:
-        """Ask the LLM to summarize the conversation and return the summary.
-
-        Args:
-            config (dict): Configuration for the agent invocation.
-
-        Returns:
-            dict: Agent response containing content, tokens, and cost.
-        """
-        if not self.agent:
-            raise RuntimeError("Agent not initialized! Call setup() first.")
-
-        result = await self.agent.ainvoke(
-            {"messages": [{"role": "user", "content": "Summarize the conversation so far."}]}, config=config
-        )
-        messages = result.get("messages", [])
-        final_message = messages[-1] if messages else None
-
-        if not final_message:
-            logger.error("No response from agent.")
-            return {"content": "", "tokens": 0, "cost": 0.0}
-
-        final_response = final_message.content
-
-        return final_response
-
     def calculate_tokens_and_cost(self, messages: list) -> tuple:
         """
         Calculates the total token usage and estimated cost for a list of messages.