Good call. Agent-created skills persist without any approval gate by default. The Curator helps long-term (archiving unused skills after 30/90 days), but between runs you're exposed. A quick ls on skills/ before any customization session is a solid habit.
@akshay_pachaar
-
Decoupling AI Models from Agentic Orchestration Layers
By
–
Exactly. Skills as self-contained units is the key design choice. It means you can swap the model underneath without rewriting your capabilities. The harness becomes the durable layer, the model becomes interchangeable.
-
Supertonic: High-Speed Open-Source On-Device TTS Engine
By
–
this TTS model generates speech 167x faster than you can hear it. Supertonic is an on-device TTS engine that runs via ONNX for cross-platform inference. – no GPU
– 31 languages
– captures every emotion
– beats ElevenLabs on speed
– runs even on a Raspberry Pi 100% open-source. -
The emergence of goal-based patterns in AI agent architectures
By
–
write /goals like acceptance criteria. /goal is now everywhere. Claude Code, Codex, Hermes, and more agents are adopting the same pattern: you set a completion condition, the agent works autonomously until a fast evaluator model confirms the condition is met. the feature is
-
Hermes AI Agent Memory Architecture Explained
By
–
the three-tier memory of Hermes agent. AI agents forgets everything when your session ends. Hermes doesn't. it has three memory layers, each at a different speed. 𝘁𝗶𝗲𝗿 𝟭: 𝘁𝘄𝗼 𝘁𝗶𝗻𝘆 𝗺𝗮𝗿𝗸𝗱𝗼𝘄𝗻 𝗳𝗶𝗹𝗲𝘀 MEMORY.md (2,200 chars) and USER.md (1,375 chars).
-
Understanding GBrain: A Self-Wiring Knowledge System for AI Agents
By
–
What actually is GBrain? (Y Combinator CEO's personal agent brain) Every agent memory tool you've seen solves a simple problem: store facts, retrieve facts. GBrain solves a different one. It gives your agent a knowledge system that wires itself, enriches itself, and compounds
-
Essential Technical Skills for AI Engineers
By
–
As an AI Engineer. Please learn: – Harness engineering, not just prompt engineering
– Prompt caching vs. semantic caching tradeoffs
– KV cache management at scale
– Speculative decoding vs quantization
– Structured output failures & fallback chains
– Evals (LLM-as-judge + human -
Impact of Token Reduction and Context Engineering on AI Models
By
–
Exactly. At 98.7% fewer tokens, you're not tuning the same system. You're running a different one. And agreed on context engineering. The model that wins isn't the smartest one. It's the one that sees only what it needs, when it needs it.
-
Claude Code AI Agent Architecture Explained
By
–
Claude Code's architecture, mapped. Calude Code is one of the most powerful agent harnessed out there, it's a lot more than "a CLI that calls claude." the actual system has six layers, and the model is just one node inside the loop. the diagram breaks down every component:
-
Token usage debate in AI prompt interfaces for 2025
By
–
The MCP vs CLI debate. For most of 2025, AI Engineers argued about it. The skeptics had real numbers: – Playwright MCP eats 13.7K tokens
– Chrome DevTools MCP eats 18K
– A 5-server setup burns 55K tokens before any work The defenders pushed back: – CLIs break on multi-tenant