#AgentLens
🇫🇷 AgentLens audit agents IA : journal enchaîné par SHA-256, entrées OpenTelemetry et MCP, démo d’altération…
🇬🇧 AgentLens AI agent audit trail: SHA-256 chained journal, OpenTelemetry and MCP inputs, tamper demo, Docker…
#AgentLens #AgentsIA #ConformiteIA #spanobserver
AgentLens audit agents IA : piste d'audit par hachage, OpenTelemetry et MCP - Span Observer
AgentLens audit agents IA : journal enchaîné par SHA-256, entrées OpenTelemetry et MCP, démo d'altération, installation Docker et limites vérifiées.
span.observer
November 15, 2026 at 7:00 PM
Wentao Hong: AgentLens: Framework-Aware Characterization and Resource Control for Heterogeneous LLM Agents https://arxiv.org/abs/2610.25231 https://arxiv.org/pdf/2610.25231 https://arxiv.org/html/2610.25231
October 20, 2026 at 6:43 AM
New arXiv work shows 10.7% of passing SWE-agent trajectories are "Lucky Passes" with regressions or blind retries, highlighting that binary pass/fail signals hide messy agent behavior. AgentLens offers…

#SWEAgents #AgentEval #DeveloperTools #AIEngineering
https://arxiv.org/abs/2605.12925
October 7, 2026 at 12:01 AM
AI-agent-friendly runtime logging for Laravel
AI-agent-friendly runtime logging for Laravel
AI-agent-friendly runtime logging for Laravel: compact, deduplicated JSON logs in agent CLI sessions, zero overhead otherwise - Dietrich2981/laravel-agentlens
laravel-news.com
September 24, 2026 at 7:00 AM
📦 agentlens/agentlens v1.1.1

AI-agent-friendly runtime logging for Laravel: compact, deduplicated JSON logs in agent CLI sessions, zero overhead otherwise.

🔗 https://github.com/Dietrich2981/laravel-agentlens
September 20, 2026 at 9:59 AM
📦 agentlens/agentlens v1.1.0

AI-agent-friendly runtime logging for Laravel: compact, deduplicated JSON logs in agent CLI sessions, zero overhead otherwise.

🔗 https://github.com/Dietrich2981/laravel-agentlens
September 20, 2026 at 9:59 AM
April 2, 2026 at 1:45 AM
Show HN: AgentLens – Open-source observability for AI agents Hi HN, I built AgentLens because debugging multi-agent systems is painful. LangSmith is cloud-only and paid. Langfuse tracks LLM calls...

Origin | Interest | Match
Show HN: AgentLens – Open-source observability for AI agents | Hacker News
news.ycombinator.com
March 1, 2026 at 10:51 AM
Show HN: AgentLens – Open-source observability and audit trail for AI agents Built AgentLens to solve a problem we kept hitting: once you deploy AI agents, you have no idea what they're actua...

Origin | Interest | Match
GitHub - amitpaz1/agentlens: Open-source observability and audit trail platform for AI agents. MCP-native, tamper-evident event logging, real-time dashboard.
Open-source observability and audit trail platform for AI agents. MCP-native, tamper-evident event logging, real-time dashboard. - amitpaz1/agentlens
github.com
February 8, 2026 at 9:28 AM
Jeonghyeon Kim, Byeongjun Joung, Junwon Lee, Joohyung Lee, Taehoon Min, Sunjae Lee: AgentLens: Adaptive Visual Modalities for Human-Agent Interaction in Mobile GUI Agents https://arxiv.org/abs/2604.20279 https://arxiv.org/pdf/2604.20279 https://arxiv.org/html/2604.20279
April 23, 2026 at 6:41 AM
Priyam Sahoo, Gaurav Mittal, Xiaomin Li, Shengjie Ma, Benjamin Steenhoek, Pingping Lin, Yu Hu: AgentLens: Revealing The Lucky Pass Problem in SWE-Agent Evaluation https://arxiv.org/abs/2605.12925 https://arxiv.org/pdf/2605.12925 https://arxiv.org/html/2605.12925
May 14, 2026 at 6:44 AM
El descubrimiento Cuando un programador le pide a un agente de inteligencia artificial que arregle un error o construya una funcion, lo importante no es solo si al final el codigo pasa las pruebas.
Este examen no pregunta si la IA resolvio el problema, sino como lo hizo: llega AgentLens - Sinaptica
El descubrimiento Cuando un programador le pide a un agente de inteligencia artificial que arregle un error o construya una funcion, lo importante no es solo si
sinapti.ca
July 10, 2026 at 2:37 AM
agentlens 0.1.44

lightweight abstractions for building agent scaffolds

Author: danielreuter
May 13, 2025 at 10:00 PM
DevTools for agents is a game changer. Debugging distributed task execution + API calls is painful without visibility. Is AgentLens logging full traces or focused on network layer?
April 2, 2026 at 1:18 AM
Podivilov, Lomshakov, Savin, Startsev, Pozharskiy, Parshin, Nikolenko: AgentLens: Production-Assessed Trajectory Reviews for Coding Agent Evaluation https://arxiv.org/abs/2607.06624 https://arxiv.org/pdf/2607.06624 https://arxiv.org/html/2607.06624
July 9, 2026 at 6:38 AM
Weidi Luo, Qiming Zhang, Yihao Quan, Mingyu Jin, Jie Cai, Chaowei Xiao, Jingcheng Niu, Zhen Xiang: AgentLens: Interpretable Safety Steering via Mechanistic Subspaces for Multi-Turn Coding Agent https://arxiv.org/abs/2606.22673 https://arxiv.org/pdf/2606.22673 https://arxiv.org/html/2606.22673
June 23, 2026 at 6:40 AM
Code agent benchmarks typically reduce runs to pass or fail. AgentLens evaluates the full trajectory: instruction following, tool usage, error recovery, and communication. Combines formal verification with LLM reviews for readable explanations. Open source.
AgentLens: Production-Assessed Trajectory Reviews for Coding Agent Evaluation
We present AgentLens, a production-assessed benchmark for interactive code agents. Most code-agent benchmarks reduce a run to a single bit -- did the task pass? -- but the people who actually use these agents experience the entire trajectory: how the agent follows instructions, uses its tools, verifies its own work, recovers from mistakes, and talks to them along the way. AgentLens evaluates that whole trajectory. It pairs formal verification, where an objective check exists, with LLM-written trajectory reviews and side-by-side comparisons, so that each run yields a readable explanation of why the score is what it is. This makes AgentLens useful for more than ranking models: we use it to diagnose model behavior, compare successive versions of our own agent, and catch product regressions in a nightly evaluation pipeline. We release the benchmark as open source at https://github.com/agent-lens/agent-lens-bench.
arxiv.org
July 11, 2026 at 3:20 AM
AI 编码代理的评估新进展:AgentLens 框架提供生产环境下的轨迹审查,旨在更全面地衡量代理表现,超越简单的成功/失败二分法。这对于优化 AI 助手在真实代码开发中的效能至关重要。
AgentLens: Production-Assessed Trajectory Reviews for Coding Agent Evaluation
AI 编码代理的评估新进展:AgentLens 框架提供生产环境下的轨迹审查,旨在更全面地衡量代理表现,超越简单的成功/失败二分法。这对于优化 AI 助手在真实代码开发中的效能至关重要。
arxiv.org
July 9, 2026 at 8:01 AM
Found a tool that helps agents capture information better.

AgentLens — agentlenshq.com

Reading comprehension layer for AI agents. Better than raw scraping. Thinking-model friendly.

Not affiliated, just found it useful for info capture problems.

Anyone tried something similar?
August 12, 2026 at 5:43 PM
🤖 AgentLens Benchmark Evaluates AI Code Agents' Full Trajectories

AgentLens, a new benchmark for interactive code agents, assesses the entire trajectory of an agent's run, not just its success or failure. This new...

#Benchmarking #GenerativeAI #OperationalEfficiency #AI #AIPulse
Read the full article →
www.synestesia.uk
July 9, 2026 at 2:36 PM
The Discovery When a programmer asks an artificial intelligence agent to fix a bug or build a function, what matters is not just whether the code passes the tests in the end.
This exam doesn't ask if the AI solved the problem, but how it did: AgentLens arrives. - Sinaptica
The Discovery When a programmer asks an artificial intelligence agent to fix a bug or build a function, what matters is not just whether the code passes the tes
sinapti.ca
July 10, 2026 at 2:37 AM
Show HN: AgentLens – Chrome DevTools for AI Agents (open-source, self-hosted)
via Dev.to

https://flarestart.com/article/show-hn-agentlens-chrome-devtools-for-ai-agents-open-source-self-hosted-20260401
#DevNews #WebDevelopment #Tutorial
Show HN: AgentLens – Chrome DevTools for AI Agents (open-source, self-hosted)
Agents are opaque. AgentLens is Chrome‑DevTools for AI agents — self‑hosted, open‑source. It traces tool calls and visualizes decision trees so you can see why an agent picked a tool. Repo:...
flarestart.com
April 1, 2026 at 6:19 PM
⬡ AgentLens: Real-Time Visualisation for Your AI Agent Pipelines
via Medium Python

https://flarestart.com/article/agentlens-real-time-visualisation-for-your-ai-agent-pipelines-20260327
#DevNews #ProgrammingLanguages
⬡ AgentLens: Real-Time Visualisation for Your AI Agent Pipelines
Stop flying blind. See exactly what your LangGraph, PydanticAI, or Agno agent is doing — as it happens. Continue reading on Medium »
flarestart.com
March 27, 2026 at 1:54 AM