proomt

Search

Search posts, papers, and topics

All posts

TemporalMason Egger15 min readintermediate

Everything in its place: Making agents write correct Temporal applications

Read on temporal.io

Summary

Use three Temporal‑specific resources to give LLM agents the right context and guardrails: the Temporal Developer Skill (curated instructions), the official SDK samples repos (executable recipes), and the Temporal Knowledge Base MCP (RAG‑powered, citation‑rich Q&A). Explicitly load the skill, add the sample directories, and wire the MCP so the agent can validate designs before emitting code. Keep…

  • Load the Temporal Developer Skill at the start of every session (e.g. `/temporal:temporal-developer`).
  • Add the language‑specific SDK samples repo to the agent’s workspace and tell the agent to reference it (`/add-dir`).
  • Connect the agent to the Temporal Knowledge Base MCP server so it can query authoritative docs and get citation links; treat “no answer” as a signal to stop hallucinating.
  • Treat the three resources as complementary: skill = condensed best‑practice guide, samples = concrete, runnable code, MCP = deep, searchable knowledge base.

LLM agents can generate plausible but incorrect Temporal code. Supplying high‑quality, curated context and a reliable source of truth dramatically reduces design bugs, mis‑use of Temporal features, and wasted token cycles, turning the agent from a noisy code generator into a productive co‑developer…

6/10
devopsaiagentssoftware engineeringtemporal

Related reading

  2. Presentation: Context Engineering at LinkedIn: How We Built an Organizational Context Layer for AI Agents with MCP

    LinkedIn built a “Contextual Agent Playbooks and Tools” layer that wraps internal services (code search, docs, feature flags, etc.) behind the open‑source Model Context Protocol (MCP). By feeding LLM‑powered coding agents the exact internal artifacts they need, the agents can diagnose incidents, generate PRs, and update incident tickets in minutes, delivering a reported 20 % productivity gain wit…

    InfoQinfoq.com28 mintalk

  4. Label your agent steps

    This post demonstrates how to use Temporal's `summary` fields for Activities, Timers, and Workflows to enhance observability. These labels make it significantly easier to debug and understand the execution flow of complex, especially agentic, workflows directly in the Temporal UI without needing to inspect full payloads.

    Temporaltemporal.io4 min

  5. Grounded Skill Synthesis from Code at Scale for Agentic Intelligence

    Code2Skill is an automated pipeline that mines popular GitHub repositories to extract verifiable, implementation‑anchored procedural “skills”. It builds a bank of ~1 M skill records (atomic ops, workflows, patterns) with provenance metadata, verifies each via blind reconstruction, and shows that augmenting LLM‑based agents with these skills yields an average 11.7% performance lift across 72 proto…

    Hugging Face Daily Papersarxiv.org1 minpaper

  6. Constraint Decay: The Fragility of LLM Agents in Backend Code Generation

    A systematic evaluation of LLM agents generating multi‑file backend code shows a sharp drop in correctness when structural constraints (framework conventions, ORM usage, API contracts) are added. Across 100 tasks in 8 Python web frameworks, assertion pass rates fall ~27 points, with data‑layer bugs (bad queries, ORM violations) driving most failures. Mid‑size models cope with minimal frameworks (…

    arXiv cs.SE (Software Engineering)arxiv.org1 minpaperHN287197