» Tag
llm
535 postsWhy Coding Agents Improve More From Harness Design Than Models
LangChain pushed a coding agent to Terminal-Bench's top 5 without changing the model — the gains came entirely from harness design.
TormentNexus AI Skill Registry Reaches 5,776 Modules
TormentNexus' AI skill registry hits 5,776 modules, enabling chainable automation for code review, Terraform generation, and database migrations.
A Decade of Vision-Language Models: Why Easy Benchmarks Mask Real Progress
A decade-long study finds vision-language model progress is real but hidden by easy benchmarks; only spatial reasoning errors remain unsolved.
CommitBrief — AI code reviews, right in your terminal
A provider-agnostic, local-first CLI that reviews your staged changes, a historic range, or a whole GitHub pull request. Zero telemetry, no server. Free and open source.
commitbrief.comVoltMem: A Volatility-Aware Memory Layer for LLM Agents
VoltMem adds domain-specific volatility priors to LLM agent memory, resolving stale-vs-stable fact conflicts that flat memory layers like Mem0 miss.
Claude Code's Prompt Cache Cost Depends on How You Launch It
Claude Code prompt-cache costs vary by launch path: measurements reveal a large fixed overhead and a subagent fan-out cost multiplier.
Synapse Lets AI Flag False Positives, Never Delete Findings
Synapse's two-model AI triage flags false positives in security scans without deleting findings, preserving gate reliability and audit trails.
Why AI Agents Forget by Design: The Memory Problem
LLM APIs are stateless by design, and the context window is not real memory. This architectural choice drives cost, latency, and consistency failures in production agents.
APC Framework Closes Authorization Gaps in Multi-Agent LLM Systems
A new authorization framework, APC, tracks delegated authority to block prompt-injection and unsafe action combinations in AI agents.
How RAG Hallucinations Poison Your Vector Database
A fintech RAG pipeline poisoned its own vector store via silent hallucinations; here's how deterministic validation fixed the problem.
CodeEraser: A Deterministic Gate Against LLM-Induced Code Drift
CodeEraser is a deterministic, non-LLM tool that blocks duplicate code and stale docs in AI-assisted codebases at write time and in CI.