Alexander Krykun

[Daniel San @dani_avila7]

Repo to reproduce in the thread:

https://t.co/MuEQuKNkF7

4 scenarios:
- noisy main context
- clean subagent
- fork subagent
- 5-level nested chain from the video

[aditya @adxtyahq]

for anyone asking where to learn this stuff:

• RAG → https://t.co/4bzbUIwV5g
• Agentic RAG → https://t.co/IotOiGmV1Y
• AI Agents → https://t.co/nEeMnVJQbk
• Multi-Agent Systems → https://t.co/pavDPVJEFj
• LangGraph → https://t.co/3miEqqFzF0
• LangGraph (code) → https://t.co/v7kxHZXqba
• MCP → https://t.co/lKawRb4etX
• Memory Systems → https://t.co/LSaT2UaPAS
• Evals → https://t.co/vxChxa1kqQ
• Context Engineering → search "Context Engineering Survey" on arXiv

and please skip the "build an ai agent in 10 minutes" videos
build something, watch it fail, then figure out why.

[Akshay 🚀 @akshay_pachaar]

Microsoft just open-sourced SkillOpt!

A framework for training agent skills like neural networks:

SkillOpt treats a plain markdown file as the trainable parameter of a frozen LLM agent, applying the same optimization discipline used in weight training: learning rates, validation gates, batch sizes, and epoch schedules.

The analogy maps precisely. The skill document is the parameter. Trajectory-derived edits are the gradient direction. An edit budget is the learning rate. A held-out split is the validation check.

Here's how it works.

A frozen model runs tasks with the current skill and produces scored trajectories. A separate optimizer model analyzes failures in minibatches, proposes structured add/delete/replace edits, and ranks them under a budget cap.

If the candidate skill improves performance on a held-out split, the edit is accepted. If not, it's rejected and stored so the optimizer avoids repeating failed changes.

The deployed output is a single best_skill. md file, typically 300 to 2,000 tokens. No weight changes, no extra inference-time calls.

The learned rules are compact and readable. These read like rules a thoughtful engineer would write after a day with the benchmark, except they were discovered automatically.

Learn more:

Paper: https://t.co/sdj5DW7t9h
GitHub: https://t.co/W3DcpBCni0

SkillOpt isn't the first system to treat skills as something you can optimize.

Hermes Agent independently built the same idea through a combination of skill_manage, Curator, and an optimization loop called GEPA that scores, mutates, and promotes skill documents across runs.

Two teams, different architectures, same conclusion: the skill file is the highest-leverage thing to optimize in a frozen-model agent.

I wrote a deep dive on how the Hermes agent works and covered all of these topics briefly.

The article is quoted below.

[Vaishnavi @_vmlops]

Harness
Engineering
A Design Guide to Claude Code

https://t.co/SZ4NcP6lDe

[Salesforce Architects @SalesforceArchs]

Whether you're designing for @salesforce Headless 360 or governing what agents are allowed to do before they do it, trust has to be designed in to your solution.
🧩

This 🆕 blog breaks down how Salesforce authenticates, authorizes, and audits external agents calling your org, and how to apply the right trust model for your agentic architecture.

Read the blog: https://t.co/mF74kNhchs

[Tom Dörr @tom_doerr]

87 design skills for Claude Code

https://t.co/O3vSxLUzd9

[How To Prompt @HowToPrompt__]

The entire RAG industry is about to get cooked.

Researchers have built a new RAG approach that:

- does not need a vector DB.
- does not embed data.
- involves no chunking.
- performs no similarity search.

It's called PageIndex. Instead of chunking your docs and stuffing them into pinecone, it builds a tree index and lets the LLM reason through it like a human reading a book.

hit 98.7% on financebench. beats every vector RAG on the leaderboard.

no embeddings. no chunking. no vector DB.

100% open source.

[Salesforce Developers @SalesforceDevs]

#TDX26 is over, but the learning doesn't stop here.

Take the TDX Quest to learn how Agentic AI is changing the game and get hands-on with the latest Agentforce innovations.

Complete by May 31, 2026 for a chance to win* a TDX Prize Box or @Salesforce cert voucher. ➡️ https://t.co/FvNfM3wtc9

*See Official Rules

[Ruben Hassid @rubenhassid]

Stop bookmarking 50 guides you'll never read.

You can skip all of it with these 15 free guides:

Claude 101: https://t.co/jw2qdIcjnh
Claude Code: https://t.co/UgE9xBXVbE
Claude Skills: https://t.co/6cHYYfjXEA
Stop prompting: https://t.co/j1LATSJiat
Claude in Excel: https://t.co/mfcXYSACWR
1M followers with AI: https://t.co/jZwxZr4ZhU
Claude for your team: https://t.co/qxlcqhf8bM
No prompt saves you: https://t.co/SDKJWylftC
AI Slides (PPT in 2026): https://t.co/L0bPMgXci6
Set up Claude Cowork: https://t.co/uWTpOI3Woc
Claude to sound like you: https://t.co/99RzxXU3p0
Claude interactive charts: https://t.co/ebCHGZqOF1
Claude as your computer: https://t.co/TxYuHPjgbV
Claude Cowork + Project: https://t.co/Q7AN9CZAbO
Set up AI before prompting: https://t.co/pE3OF72A04
___

1. Save this list for later (three dots, top right).
2. Share it with a friend by ♻️ reposting this image.
3. Subscribe to my free newsletter: https://t.co/psB7XxB2Y4.

[Ronin @DeRonin_]

10 GitHub repos to spend 60-90% less tokens in Claude Code:

1. RTK (Rust Token Killer)

CLI proxy that filters terminal output before it hits your context

- 60-90% reduction on common dev commands
- one binary, zero dependencies
- works with Claude Code, Cursor, Copilot

Repo: https://t.co/WayvpBtyBH

2. Context Mode

Sandboxes raw tool output into SQLite instead of dumping it into context

- 98% context reduction on Playwright, GitHub, logs
- only clean summaries enter your conversation
- works as Claude Code plugin

Repo: https://t.co/YNbFIGQz7X

3. code-review-graph

Local knowledge graph that maps your codebase with Tree-sitter

- Claude reads only what matters, not the entire repo
- 49x token reduction on large monorepos
- 6.8x on average reviews

Repo: https://t.co/9gIzmAWN12

4. Token Savior

MCP server that navigates code by symbols, not full files

- 97% reduction on code navigation
- persistent memory across sessions
- 69 tools, zero external deps

Repo: https://t.co/OtvhrMgGWh

5. Caveman Claude

makes Claude talk like a caveman to cut output tokens

- 65-75% output reduction
- one-line install
- keeps full technical accuracy

Repo: https://t.co/onBeghTyfH

6. claude-token-efficient

one CLAUDE.md file that keeps responses terse

- drop-in, no code changes
- reduces output verbosity on heavy workflows
- best for output-heavy sessions

Repo: https://t.co/j6MKo9klQe

7. token-optimizer-mcp

MCP server with caching, compression, and smart tool intelligence

- 95%+ token reduction through intelligent caching
- compresses repeated tool outputs

Repo: https://t.co/0jIVQ4ANls

8. claude-token-optimizer

reusable setup prompts for optimizing any project

- 90% token savings in 5 minutes
- reduces doc token usage from 11K to 1.3K

Repo: https://t.co/puil9WwFGB

9. token-optimizer

finds ghost tokens that silently eat your context

- survives compaction without losing quality
- fixes context quality decay

Repo: https://t.co/92G8e4yeGq

10. claude-context (by Zilliz)

code search MCP that makes your entire codebase the context

- ~40% reduction with equivalent retrieval quality
- hybrid BM25 + dense vector search

Repo: https://t.co/yjfiQOSy15

[ how to stack them ]:

you don't need all 10. pick 2-3 based on your workflow:

> heavy terminal output? RTK
> big codebase? code-review-graph + Token Savior
> lots of MCP servers? Context Mode
> quick fix? Caveman + claude-token-efficient

most people are burning tokens without knowing it

run /context in a fresh session and see how much is gone before you even type a word

your pocket will thank me later :<)

[Salesforce @salesforce]

The agentic economy has landed.

Meet AgentExchange—one marketplace for trusted AI agents, apps, and @SlackHQ.

10K+ agents &amp; tools. 160K+ enterprise customers.

Builders &amp; ISVs: distribute, monetize, and scale.

Backed by a $50M Builders Initiative. See it live at #TDX26.

[Salesforce Developers @SalesforceDevs]

📣 Introducing a brand new sample app: @Salesforce Multi-Framework Recipes.

Dive into a collection of easy-to-digest code examples for building modern web apps on Salesforce.

Explore the recipes today. 👉 https://t.co/AZaIkv1vZd

[Headless 360 platform @SalesforceforIT]

📣 Innovate with total flexibility with Salesforce Multi-framework. #TDX26

Build native Salesforce apps with React on the Agentforce 360 Platform, with built-in authentication, security, and governance.

Learn about the launch in the blog. https://t.co/SArnLxVcFi

[divyansh tiwari @DivyanshT91162]

Best GitHub repos for Claude Code that will 10x your next project in 2026:

1. Claude Skills
https://t.co/C0mQ31YOHt…

2. Claude Code Plugins + Skills
https://t.co/tEvfqFF5zv…

3. Awesome Claude Skills
https://t.co/jGnoJ4ECqU…

4. Awesome Claude Skills
https://t.co/CUNRb5NOI9…

5. Repomix
https://t.co/LXQmjCslD9…

6. Claude Session Restore
https://t.co/aMDjZFXUyS…

7. Awesome Claude Code Toolkit
https://t.co/tcffwODGV4…

8. Awesome Claude Code
https://t.co/bp1tG9mDT8…

9. Claude Code Best Practice
https://t.co/NGUp2Cfct9…

10. Superpowers
https://t.co/sL8s2qE6Cr…

[Hasan Toor @hasantoxr]

10 Best open source repos to try this weekend:

1. Clawless: https://t.co/kQU9gHtxEZ

2. Paperclip: https://t.co/YlOizM7jW7

3. TimesFM:
https://t.co/cZOrfMGiVr

4. Lark CLI:
https://t.co/loPCwGRLiD

5. ST3GG:
https://t.co/8Jw73IbyiZ

6. OpenClaude:
https://t.co/fGh1hsbafT

7. Agent Orchestrator:
https://t.co/ZZ0GZNyjdD

8. MiroFish:
https://t.co/yX2D7w5TLJ

9. Hermes Agent:
https://t.co/XSTGgrkvRP

10. Oh my claudecode:
https://t.co/zF8i027Jra

[Dhairya @dkare1009]

Claude Code Project Structure (Quick Guide)

Most projects start with just CLAUDE.md.
That’s why they stay basic.

Here’s the structure that turns Claude into an engineering partner 👇

📌 Core Files
• CLAUDE.md → project rules & context
• CLAUDE.local.md → personal overrides
• mcp.json → connects tools (GitHub, Slack, DBs)
• claude/settings.json → permissions + models

📂 claude/ (real power)
• rules/ → coding + testing standards
• commands/ → reusable workflows (/review, /fix)
• skills/ → auto-loaded expertise (on-demand)
• agents/ → specialized sub-agents
• hooks/ → automation + guardrails

💡 Better structure = better output

Set it once → every session improves.

What does your setup look like?

Save 💾 • Follow for AI workflows 🚀

[Hasan Toor @hasantoxr]

Best GitHub repos for Claude code that will 10x your next project:

1. Superpowers
https://t.co/U5Y4BK9Lap

2. Awesome Claude Code
https://t.co/qcgoxU3Up2

3. GSD (Get Shit Done)
https://t.co/WfAhllWnTR

4. Claude Mem
https://t.co/XLQpwdnIWN

5. UI UX Pro Max
https://t.co/aQtGjMzKus

6. n8n-MCP
https://t.co/7le1aluZXH

7. Obsidian Skills
https://t.co/MUaoyUnasw

8. LightRAG
https://t.co/ye8z4UqaMc

9. Everything Claude Code
https://t.co/OAU9JE46Uz

[Tech with Mak @techNmak]

I thought I was using OpenClaw well.

Then I watched this.

This is the OpenClaw masterclass.

[kepano @kepano]

1. install Obsidian 1.12
2. enable CLI
3. now OpenClaw, OpenCode, Claude Code, Codex, or any other agent can use Obsidian

[Salesforce Architects @SalesforceArchs]

How do architects translate complex requirements to select the right integration pattern? 💭

Join us Feb. 19 for episode two of Think Like an Architect as experts apply the What/How/Why to a real-world integration scenario.

Register today: https://t.co/AHVCvL7I2G