backpass: Gradient Descent for Agent Memory and AGENTS.md

Repositorykunchenguid/backpassTool

backpass is an open-source CLI tool that parses local agent transcripts to propose evidence-backed updates to project memory files like AGENTS.md or CLAUDE.md.

Key points

  1. Conceptual approach: Treats memory surfaces (AGENTS.md, skills) as weights, agent runs as forward passes, and on-disk transcripts as loss signals.
  2. Evidence-gated edits: Additions, rewrites, or deletions require verbatim quote evidence sourced from at least two distinct sessions.
  3. Local-first design: Reads transcript stores across seven agent harnesses directly from disk, redacting obvious secrets locally before any processing.
  4. Human in the loop: Analysis never modifies files automatically; developers review and approve changes via backpass apply.
  5. Runtime requirements: Built with JavaScript and requires Node.js >= 22.5 alongside acpx in the PATH.

Why it matters

Instruction files like AGENTS.md or CLAUDE.md are typically updated manually after developer frustration. backpass systematises this feedback loop by turning session logs into iterative, budget-conscious memory improvements.

Analysis

By framing prompt and skill maintenance as bounded gradient steps, backpass prevents hallucinated rewrites and protects memory context from bloat. The requirement of dual-session corroboration adds pragmatic guardrails against overfitting to single run anomalies. With solid star growth and automated release pipelines, it offers a structured workflow for AI-assisted engineering teams.

Mentioned links

Found in

Summary generated by AI (Gemini 3.8 Flash, 27 September 2026). It may contain errors – the original source is authoritative.

Related entries

Similar in content, found by the AI search.

  • X post:Kun Chen

    backpass: Gradient Descent for AGENTS.md and CLAUDE.md

    Kun Chen recommends against immediately rewriting CLAUDE.md or AGENTS.md when switching to newly released AI models. Instead, teams should keep their existing setup initially and use the open-source CLI tool backpass to propose evidence-backed refinements derived from session transcripts.

    633bookmarks57.2kviews

    AI & LLMs· Tool

  • Article:Kun Chen

    Treating AGENTS.md as a Neural Net: Kun Chen Introduces Backpass

    Kun Chen argues that project-level memory files like AGENTS.md or CLAUDE.md should be treated as neural network weights. Rather than manually appending rules based on anecdotes, he applies a gradient-descent loop driven by actual session transcripts, implemented in the open-source CLI tool backpass.

    AI & LLMs· Tool

  • Repository:rohitg00/agentmemory

    agentmemory: Persistent Memory for AI Coding Agents

    agentmemory is an open-source TypeScript solution providing persistent memory for coding agents and MCP clients. Powered by the iii engine, it builds on Andrej Karpathy's LLM Wiki pattern by introducing confidence scoring, lifecycle management, knowledge graphs, and hybrid search.

    28kstarsTypeScript

    AI & LLMs· Tool

  • X post:Kun Chen

    Optimising AI Agent Skills via Backward Pass over Transcripts

    Kun Chen advocates systematically optimising AI assistant memory and skill files using a 'backward pass'. Instead of reacting to one-off anecdotes, large samples of session transcripts should be analysed to distill losses and update skills systematically.

    378bookmarks9.5kviews

    AI & LLMs· Discussion

  • Repository:thedotmack/claude-mem

    claude-mem: Persistent Cross-Session Memory for AI Agents

    The open-source project claude-mem (now also branded as Grok Mem) captures agent activity during active sessions, compresses it with AI, and injects relevant context back into future runs.

    93.1kstarsJavaScript

    AI & LLMs· Tool

Back to the knowledge base

Let's talk about your project

Get in touch

Locations

  • Mattersburg
    Johann Nepomuk Bergerstraße 7/2/14
    7210 Mattersburg, Austria
  • Vienna
    Ungargasse 64-66/3/404
    1030 Wien, Austria

Parts of this content were created with the assistance of AI.