{"slug": "coding-agents-are-wasting-too-many-tokens-rediscovering-things", "title": "Coding agents are wasting too many tokens rediscovering things", "summary": "Decispher launched a persistent memory layer for coding agents that cuts token usage by 38×, according to its LongMemEval benchmarks, by organizing context from GitHub, documentation, and engineering platforms into reusable context units. The system, which includes a Context Engine, Memory Plane, and Worker Agent, reported 89% accuracy on the oracle split using GPT-4o-mini and integrates with Cursor, Claude, Codex, and Grok Build.", "body_md": "# Coding agents are wasting too many tokens rediscovering things\n\nDecispher is trying to solve this by building a persistent memory layer specifically for engineering workflows. Instead of just feeding an LLM a massive dump of files, it acts as a middle layer that pulls context from fragmented sources and organizes it into something an agent can actually use.\n\n## The architecture of an engineering memory layer\n\nThe system is split into three distinct functional blocks, which is a smart way to handle the \"context window\" problem without bloating every single request.\n\n**The Context Engine:** This is the retrieval part. It doesn't just look at code; it pulls from GitHub, engineering platforms, and documentation to create \"context units.\" For instance, if an agent is working on a payment module, the engine gathers the relevant PRs, architectural decision records (ADRs), and implementation history. They also have a feature called \"Branch Story\" that turns an AI's execution path (Prompt → plan → actions → result) into a structured handoff for the human reviewer on the PR.**The Memory Plane:** This handles long-term storage at the user, team, and project levels. It's meant to hold things like coding conventions or specific backend preferences. According to their LongMemEval benchmarks, they are seeing a 38× median token reduction, which is massive for anyone worried about LLM inference costs.**The Worker Agent:** This is the autonomous piece. It can pull tasks directly from Jira or Slack, fetch the necessary context, and—crucially—ask a human for clarification instead of hallucinating a solution when the context is thin.\n\n## Implementation and security details\n\nIf you want to test the Context Engine, the setup looks like a standard CLI workflow:\n\n```\nnpx decispher init\n# This connects your repo and configures agent integration\n\nnpx decispher link\n# This links your Decispher account to the repo\n```\n\nThe tool is designed to work with MCP-compatible agents, and they've built specific integrations for [Cursor](/en/tags/cursor/), Claude, Codex, and Grok Build. There is also a VS Code extension for managing handoffs.\n\nFrom a security standpoint, they've addressed the typical \"agent in my repo\" fears. The Context Engine doesn't clone your whole source tree; it interacts via the GitHub API. The Worker Agent runs in an isolated sandbox with no network access except through a strictly allowlisted proxy, and those sandboxes are wiped immediately after the run.\n\nThe heavy lifting here is the retrieval quality. They reported 89% accuracy on the oracle split using GPT-4o-mini as the extractor/reader. If they can actually maintain that level of precision while cutting token usage by 38x, this moves from a \"nice-to-have\" to a core part of the AI workflow.\n\n[Next The US government just seized an Anthropic stake linked to the →](/en/news/8355/)\n\n## All Replies （4）\n\n`context.md`\n\nfile with architecture notes; it saves a ton of tokens.", "url": "https://wpnews.pro/news/coding-agents-are-wasting-too-many-tokens-rediscovering-things", "canonical_source": "https://promptcube3.com/en/news/8358/", "published_at": "2026-08-31 13:41:20+00:00", "updated_at": "2026-08-31 13:54:27.528639+00:00", "lang": "en", "topics": ["ai-tools", "ai-infrastructure", "developer-tools", "machine-learning"], "entities": ["Decispher", "GitHub", "Cursor", "Claude", "Codex", "Grok Build", "GPT-4o-mini", "LongMemEval"], "alternates": {"html": "https://wpnews.pro/news/coding-agents-are-wasting-too-many-tokens-rediscovering-things", "markdown": "https://wpnews.pro/news/coding-agents-are-wasting-too-many-tokens-rediscovering-things.md", "text": "https://wpnews.pro/news/coding-agents-are-wasting-too-many-tokens-rediscovering-things.txt", "jsonld": "https://wpnews.pro/news/coding-agents-are-wasting-too-many-tokens-rediscovering-things.jsonld"}}