Tag

#agentic-coding

15 posts

Security & PrivacyOctober 26, 202614 min
Claude Code safety_event: What It Logs, Tested
Claude Code 2.1.307 added a safety_event OpenTelemetry event. I ran 2.1.306, 2.1.307 and 2.1.310 to see which refusals it logs and which it skips in -p runs.
Security & PrivacyOctober 22, 202615 min
Claude Code MCP Permission Rules: 2.1.299 Tested
Claude Code 2.1.299 changed which MCP servers an allow or deny rule covers. I ran 2.1.296, 2.1.299 and 2.1.307 with servers named a and a__b. Here is what ran.
Security & PrivacyOctober 21, 202616 min
Claude Code updatedInput Hooks: 2.1.300 Change Tested
Claude Code 2.1.300 validates updatedInput from PreToolUse hooks. I compared 2.1.299 with 2.1.300 and 2.1.306 on unknown keys, bad types, paths and Edit.
Security & PrivacyOctober 13, 202620 min
Claude Code Hook Failures: onFailure Block Tested
Claude Code hooks fail open on most events. I tested onFailure block across four builds with cases like missing scripts, exit 1, timeouts and schema errors.
Tooling & ProductivityOctober 9, 202635 min
Claude Code Mods: 10 Tested Use Cases (2026)
Claude Code mods explained with 10 tested use cases: guard risky commands, mask secrets, add commands, panes and tools. Every example passes claude plugin test.
Security & PrivacyOctober 5, 202613 min
Claude Code Mod to Block .env Reads: Tested (2026)
Build a Claude Code mod to block .env reads. It denies Read, Edit, Write and Bash calls on secret files, passes 5 tests, and leaks on 5 of 14 shell commands.
NewsSeptember 14, 202634 min
GitHub HydraFusion: Multi-Model Copilot Routing 2026
GitHub's HydraFusion can route one coding prompt across several models at runtime. Its own table shows cost down on all three benchmarks and quality up on one.
NewsSeptember 4, 202619 min
Coding Agents vs SaaS: The 2026 Build-vs-Buy Shift
32% of respondents told McKinsey their org skipped a software purchase because coding agents could build it. What that number measures, and what it does not.
NewsJuly 22, 202613 min
Kimi K3 vs Fable 5, GPT-5.6, Grok: 2026 Benchmarks
Kimi K3 is the first open 3T-class model. See how Moonshot's launch benchmarks stack up against Claude Fable 5, GPT-5.6 Sol, Grok 4.5, and Gemini in 2026.
NewsJuly 1, 202610 min
Claude Sonnet 5: Cheaper Agents, Near-Opus Power (2026)
Claude Sonnet 5 brings near-Opus 4.8 agentic coding at $2/$10 per million tokens through August. See the pricing, the tokenizer catch, and the system card.
AI & Machine LearningJune 13, 202610 min
Kimi K2.7-Code: Open Weights, First-Party Numbers (2026)
Moonshot's Kimi K2.7-Code is a 1T open-weight coding model with cheap API pricing, but every launch benchmark is first-party. Here's what's actually verified.
AI & Machine LearningMay 28, 202619 min
DeepSWE: AI Coding Benchmark Catches Claude Cheating in 2026
Datacurve's DeepSWE coding benchmark crowns GPT-5.5 at 70%, catches Claude Opus 4.7 reading gold commits from .git history, and exposes SWE-Bench Pro flaws.

Every post in #agentic-coding (15)