Author

Code Mala Tang

Read source code together, write articles together, and enjoy spicy hot pot together.

521

Articles

Likes

4.7k

Views

Comments

Latest from Code Mala Tang

100 recent articles max

Code Mala Tang

Oct 22, 2026 · Information Security

AI Discovers Critical PHP RCE in 5 Days; Patch Coordination Takes 58

The article analyzes the PHP2Shell vulnerabilities (CVE-2026-109618/19), discovered by AI in 5 days but requiring 58 days to patch due to coordination across PHP, distributions, cloud providers, and WAF vendors, highlighting the asymmetry between AI-accelerated vulnerability discovery and human-driven remediation.

AI security researchCVE-2026-109618CVE-2026-109619

0 likes · 10 min read

AI Discovers Critical PHP RCE in 5 Days; Patch Coordination Takes 58

Code Mala Tang

Sep 7, 2026 · Cloud Native

Kubernetes Resource Contracts: How requests and limits Control Pod Scheduling, Throttling, and Eviction

This article explains how Kubernetes requests and limits act as a contract between workloads and the scheduler, detailing their distinct roles, the divergent consequences of CPU throttling versus memory OOMKills, the three QoS classes that determine eviction priority, and how LimitRange and ResourceQuota enforce resource policies at the namespace level.

CPU throttlingKubernetesLimitRange

0 likes · 10 min read

Kubernetes Resource Contracts: How requests and limits Control Pod Scheduling, Throttling, and Eviction

Code Mala Tang

Sep 6, 2026 · Cloud Native

Kubernetes Volume Mounts: Hot Reload, subPath Pitfalls & Projected Volumes

This article explains Kubernetes volume mounting for ConfigMaps and Secrets, detailing how directory mounts achieve hot reloads via atomic symlink switching, why subPath mounts never update, and how projected volumes merge multiple configuration sources into a single directory.

ConfigMapHot ReloadKubernetes

0 likes · 10 min read

Kubernetes Volume Mounts: Hot Reload, subPath Pitfalls & Projected Volumes

Code Mala Tang

Sep 1, 2026 · Industry Insights

AI Agent Infrastructure Is Just 1996 Linux Sysadmin Practices Rebranded

The article maps modern AI Agent terminology — identity isolation, least privilege, sandbox, runtime, scheduler, observability, human-in-the-loop, secure execution, self-healing — to decades-old Linux concepts like user permissions, chmod, directories, sudo, systemd, cron, journalctl, SSH, Docker, and process supervision, arguing that Linux veterans already manage AI agents as just another untrusted user.

AI AgentsContainerizationDevOps

0 likes · 3 min read

AI Agent Infrastructure Is Just 1996 Linux Sysadmin Practices Rebranded

Code Mala Tang

Aug 18, 2026 · Artificial Intelligence

How Claude’s Text Watermark Works: Anthropic’s Technical Explanation

Anthropic explains that Claude’s text watermark, derived from DeepMind’s SynthID‑Text and Scott Aaronson’s 2022 concept, alters the model’s token‑selection randomness with a secret key to embed statistical traces without degrading quality, while outlining its detection limits, code‑generation nuances, privacy properties, and industry rollout.

AI model detectionAnthropicClaude

0 likes · 6 min read

How Claude’s Text Watermark Works: Anthropic’s Technical Explanation

Code Mala Tang

Aug 3, 2026 · Backend Development

From Missing dingtalkid to One‑Click Agent: Building DingTalk Robot Jump Links

This guide explains how to capture the hidden $:LWCP_v1:$‑prefixed dingtalkid from DingTalk robot callbacks using a minimal Python HTTP service, then construct a dingtalk://jump‑robot link that pre‑fills a query, enabling users to launch an internal Agent with a single click.

CallbackDingTalkPython

0 likes · 12 min read

From Missing dingtalkid to One‑Click Agent: Building DingTalk Robot Jump Links

Code Mala Tang

Jul 28, 2026 · Cloud Native

What’s New in MCP Protocol’s Fifth Major Release (2026‑07‑28)?

The MCP protocol’s fifth major version switches from a stateful bidirectional model to a stateless request/response design, adding new headers, auth hardening, deprecating legacy features, updating SDKs, and gaining broad cloud and tool support, while outlining migration steps for production users.

MCPSDKServerless

0 likes · 5 min read

What’s New in MCP Protocol’s Fifth Major Release (2026‑07‑28)?

Code Mala Tang

Jul 16, 2026 · Artificial Intelligence

How Hermes Agent’s Self‑Evolution Framework Works: DSPy Diagnoses, GEPA Evaluates

The article dissects the open‑source hermes‑agent‑self‑evolution repository, explaining how DSPy leverages large‑model APIs to diagnose skill failures and rewrite SKILL.md, while GEPA orchestrates mutation, evaluation, and Pareto‑based selection to iteratively improve Hermes Agent skills offline.

AI AgentsDSPyGEPA

0 likes · 11 min read

How Hermes Agent’s Self‑Evolution Framework Works: DSPy Diagnoses, GEPA Evaluates

Code Mala Tang

Jul 9, 2026 · Artificial Intelligence

Writing Effective Agent Skills: Human‑Crafted Boost +16.2 pp, AI‑Generated Drop –1.3 pp

An empirical benchmark of 8,408 runs across 84 tasks shows that manually designed Agent Skills improve performance by up to 16.2 percentage points, while AI‑generated skills can hurt results, and distills six actionable conclusions for crafting useful, focused, and test‑validated skills.

AI AgentsAgent SkillsDomain Adaptation

0 likes · 7 min read

Writing Effective Agent Skills: Human‑Crafted Boost +16.2 pp, AI‑Generated Drop –1.3 pp

Code Mala Tang

Jul 4, 2026 · Artificial Intelligence

How an AI Assistant Self‑Learns Skills and Tops GitHub Stars in 60 Days

In just 60 days the OpenClaw AI assistant, which writes and refines its own skills on a local laptop, amassed over 250,000 GitHub stars—outpacing React and Linux—while offering a self‑improving, extensible agent framework and highlighting associated security considerations.

AI AgentsOpenClawlocal AI

0 likes · 12 min read

How an AI Assistant Self‑Learns Skills and Tops GitHub Stars in 60 Days