Chat agent
A knowledge base, conversation memory and an /ask route over Claude, with sentence embeddings for retrieval.
One app.py with 3 tables, 1 view, 5 computed columns, 8 HTTP routes. Needs pixeltable[serve]>=0.7.16.
app.py
"""pxt schema update app.py agent && pxt service update app.py agentExport ANTHROPIC_API_KEY before the first pxt command; the service inherits the daemon's environment.Exported it later? The CLI refuses with STALE_CONFIG: run pxt daemon restart, then pxt service update app.py agent."""# ruff: noqa: F821from datetime import datetimefrom typing import Anyimport pixeltable as pxtimport pixeltable.functions as pxtfimport pydanticfrom fastapi import HTTPExceptionfrom pixeltable.functions.anthropic import messagesfrom pixeltable.functions.huggingface import sentence_transformerfrom pixeltable.serving import FastAPIRouterTableModel = pxt.model_base()embed_fn = sentence_transformer.using(model_id="all-MiniLM-L6-v2")_DEFAULT_SYSTEM = "You are a helpful assistant with a knowledge base and conversation memory. Be concise and accurate."@pxt.udfdef assemble_context(prompt: str,memory_context: list[dict[str, Any]] | None,knowledge_context: list[Any] | None,) -> str:parts = [f"QUESTION:\n{prompt}"]if knowledge_context:items = [item.get("text", str(item)) if isinstance(item, dict) else str(item) for item in knowledge_context]parts.append("KNOWLEDGE:\n" + "\n".join(f"- {t}" for t in items if t))if memory_context:lines = [f"- [{m.get('role', '?')}] {m.get('content', '')[:200]}" for m in memory_context]parts.append("MEMORY:\n" + "\n".join(lines))return "\n\n".join(parts)class Knowledge(TableModel, name="knowledge"):body: pxt.Stringtitle: pxt.Stringsource: pxt.Stringuuid = pxt.Column(value=pxtf.uuid.uuid7(), primary_key=True)timestamp: pxt.Timestamp | Noneclass Sentences(TableModel,name="sentences",base=Knowledge,iterator=pxtf.string.string_splitter(Knowledge.body, separators="sentence"),):__indexes__ = [pxt.EmbeddingIndex(text, embedding=embed_fn, name="knowledge_embed")]class Conversations(TableModel, name="conversations"):role: pxt.Stringcontent: pxt.Stringconversation_id: pxt.Stringuser_id: pxt.Stringuuid = pxt.Column(value=pxtf.uuid.uuid7(), primary_key=True)timestamp: pxt.Timestamp | None__indexes__ = [pxt.EmbeddingIndex(content, embedding=embed_fn, name="conversations_embed")]@pxt.querydef search_knowledge(query_text: str, limit: int = 10) -> pxt.Query:sim = Sentences.text.similarity(string=query_text)return Sentences.where(sim > 0.3).order_by(sim, asc=False).select(Sentences.text, score=sim).limit(limit)@pxt.querydef recall_memory(query_text: str, limit: int = 10) -> pxt.Query:sim = Conversations.content.similarity(string=query_text)return (Conversations.where(sim > 0.5).order_by(sim, asc=False).select(role=Conversations.role,content=Conversations.content,conversation_id=Conversations.conversation_id,score=sim,).limit(limit))@pxt.querydef get_history(conversation_id: str, limit: int = 10) -> pxt.Query:return (Conversations.where(Conversations.conversation_id == conversation_id).order_by(Conversations.timestamp, asc=False).select(role=Conversations.role, content=Conversations.content, timestamp=Conversations.timestamp).limit(limit))class Agent(TableModel, name="agent"):prompt: pxt.Stringconversation_id: pxt.Stringsystem_prompt: pxt.Stringmax_tokens: pxt.Intuuid = pxt.Column(value=pxtf.uuid.uuid7(), primary_key=True)timestamp: pxt.Timestamp | Nonememory_context = recall_memory(prompt)knowledge_context = search_knowledge(prompt)context = assemble_context(prompt, memory_context, knowledge_context)final_response = messages(model="claude-sonnet-5",messages=[{"role": "user", "content": context}],max_tokens=max_tokens,model_kwargs={"system": system_prompt},)answer = final_response.content[0].text.astype(pxt.String)api = FastAPIRouter(name="agent", prefix="/api")api.add_insert_route(Knowledge, path="/knowledge", inputs=[Knowledge.body, Knowledge.title, Knowledge.source], outputs=[Knowledge.uuid])api.add_insert_route(Conversations,path="/conversations",inputs=[Conversations.role, Conversations.content, Conversations.conversation_id, Conversations.user_id],outputs=[Conversations.uuid],)api.add_query_route(path="/knowledge/search", query=search_knowledge, method="get")api.add_query_route(path="/memory/search", query=recall_memory, method="get")api.add_query_route(path="/history", query=get_history, method="get")api.add_delete_route(Knowledge, path="/delete/knowledge")api.add_delete_route(Conversations, path="/delete/conversation")# Models are bound to their tables by `pxt service` (at the TARGET being served, local or pxt://).# From plain Python, call TableModel.bind_all("agent") after `import app`.def _append_turns(question: str, answer: str, conversation_id: str, ts: datetime) -> None:Conversations.insert([{"role": "user","content": question,"conversation_id": conversation_id,"user_id": "default","timestamp": ts,},{"role": "assistant","content": answer,"conversation_id": conversation_id,"user_id": "system","timestamp": datetime.now(),},])def ask(question: str,conversation_id: str = "default",*,system_prompt: str | None = None,max_tokens: int = 1024,) -> str:if not Agent.is_bound:raise RuntimeError("Models are not bound to tables. Run `pxt schema update app.py agent`, ""then `TableModel.bind_all('agent')` before calling ask() from plain Python.")ts = datetime.now()status = Agent.insert([{"prompt": question,"conversation_id": conversation_id,"system_prompt": system_prompt or _DEFAULT_SYSTEM,"max_tokens": max_tokens,"timestamp": ts,}],return_rows=True,)if not status.rows:return "Error: no response generated."answer = status.rows[0].get("answer") or "Error: no answer in response."_append_turns(question, answer, conversation_id, ts)return answerclass AskRequest(pydantic.BaseModel):prompt: strconversation_id: str = "default"system_prompt: str | None = Nonemax_tokens: int = 1024class AskResponse(pydantic.BaseModel):answer: str# Hand-written rather than declared: /ask orchestrates an insert plus two follow-up writes.# FastAPIRouter subclasses fastapi.APIRouter so this is served normally, but it is not part of# service_spec(), so `pxt service diff` and `pxt service list` will not mention it.@api.post("/ask")def ask_http(body: AskRequest) -> AskResponse:try:return AskResponse(answer=ask(body.prompt,body.conversation_id,system_prompt=body.system_prompt,max_tokens=body.max_tokens,))except pxt.Error as exc:raise HTTPException(status_code=exc.http_status, detail=exc.to_dict()) from exc
What app.py declares
Drawn from app.py at 15824a3; each row links to its line.
- runs on insert
- stored
- index
- route
- promptconversation_idsystem_promptmax_tokensuuidtimestamp
- memory_context = recall_memory(prompt)L107
- knowledge_context = search_knowledge(prompt)L108
- context = assemble_context(prompt, memory_context, knowledge_context)L109
- final_response = messages(model='claude-sonnet-5', messages=[{'role': 'user', 'content': context}], max_...L110
- answer = final_response.content[0].text.astype(pxt.String)L116
serves agent
- POST/api/knowledgeinsert, knowledgetakes body, title, source; returns uuidL120
- POST/api/conversationsinsert, conversationstakes role, content, conversation_id, user_id; returns uuidL123
- GET/api/knowledge/searchquery, search_knowledgetakes query_text, limitL129
- GET/api/memory/searchquery, recall_memorytakes query_text, limitL130
- GET/api/historyquery, get_historytakes conversation_id, limitL131
- POST/api/delete/knowledgedelete, knowledgeL132
- POST/api/delete/conversationdelete, conversationsL133
- POST/api/askhandlerL205
Run it
Its LLM calls need ANTHROPIC_API_KEY. Export it before the first pxt command; on Cloud, store it with pxt secret set.
Needs uv. The routes then answer on port 8000.
terminal
git clone https://github.com/pixeltable/pixeltable-starter-kit &&cd pixeltable-starter-kit &&git checkout 15824a3f20f3c67bd355c75471bbdecd8de586ec &&cd chat-agent &&uv sync &&export ANTHROPIC_API_KEY="${ANTHROPIC_API_KEY:?Set ANTHROPIC_API_KEY before running this app}" &&uv run pxt schema update app.py agent &&uv run pxt service run app.py agent --port 8000
Related apps
- Apiary AdvisorRAG + tool-calling beekeeping advisor: custom embedding index, table-as-UDF agent pipeline, expr_udf mite mathShares: Embedding indexes, Views and iterators, LLM calls, UDFs
- Research Agent MemoryDurable, searchable long-term memory for an AI research assistant: per-user semantic recall, retrieval_udf lookups, tool calling and memory consolidation with custom aggregatesShares: Embedding indexes, LLM calls, UDFs
- Video searchVideo frames extracted at one per second and indexed with CLIP, searchable by text or by image.Shares: Embedding indexes, Views and iterators