New Project? New CC?
Want to back the stream? The «داعم» membership on YouTube is EGP 60 a month: badges, emoji, and I answer your questions first on stream. Become a supporter
The gist
The stream centers on a vision for Mission Control as a unified layer for monitoring agents and sub-agents, routing work by goals and available capacity, and moving from small tasks toward longer, more autonomous builds. Luqman discusses preparing PRDs, ERDs, system design, infrastructure, and access in advance while keeping humans in the loop and adding code, security, and Playwright checks before merges; however, the design, integrations, and full automation remained exploratory, with no confirmed completion. Audience questions covered choosing tools such as Codex and Claude, context and usage management, learning Python versus AI engineering, and reverse engineering with its legal risks. Practical discussions also addressed product distribution and audience reach, email-deliverability testing, AEO for Shopify stores, interface improvement, and repository knowledge. A recurring attributed view is that focus, quality, distribution, and reviewing AI-produced work matter more than maximizing project or content volume, while several technical, legal, and experimental outcomes remained unresolved.
103 moments from the stream
- 0:51:08
Reviewing a UI with a Design-Taste Skill
An unspecified website UI issue prompts a check for an installed design-taste skill and consideration of using it; the page does not record the outcome of that review.
- 0:52:11
A Searchable Archive of Streams
A proposed section would let users search a term across the stream archive and find the recordings and relevant moments where it appeared, while seeking a practical way to reduce indexing cost.
- 0:54:38
Testing a Website Redesign in a Temporary Copy
A suggested workflow is to download and run the site, perform the redesign in a temporary copy for review, and make it the main version only if it is approved.
- 0:55:06
From Task Management to Goal-Driven Missions
The Mission Control vision replaces tactical task lists with larger missions tied to explicit goals and executed by groups of agents. It requires defining the technical stack and required access in advance, while keeping a human in the loop for sensitive setup such as accounts, APIs, and secrets. Practical output review and correction of minor defects remain part of oversight.
- 0:58:13
Managing Work as a Mission with Sub-agents
The discussion proposes giving an agent a broad “mission” rather than a narrow task, allowing it to coordinate sub-agents and execution when it has the necessary access and resources.
- 0:59:26
Previewing a Design Skill’s Effect on a Website
A temporary website version is proposed to visually compare the effect of a design-improvement skill. The speaker remains cautious about how impressive such tools are, noting frequent use of Impeccable while leaving the relative merits of the two skills unresolved.
- 1:03:34
Mission Control for Project Building
Mission Control is described as a potentially workflow-changing tool for building projects, distinct from managing systems or work that already exists. Its operating details remain incomplete in this excerpt.
- 1:04:06
What Is AEO for Shopify Stores?
AEO is explained as helping Shopify stores appear in AI-platform answers and recommendations when users ask where to buy products, in a role analogous to SEO for traditional search results.
- 1:04:47
A Repeated Codex Malware Warning
A warning repeatedly reported that Codex was not opened because it contained malware, prompting questions about background activity and whether a recurring job might explain it. No diagnosis or confirmed cause is established here.
- 1:07:14
Nexus as a Project-Isolated Local MCP Gateway
Nexus is presented as a local gateway between coding agents and services: services are linked once, while accounts, resources, and credentials remain separated by project. Its proposed value is letting an agent reach MCP integrations through Nexus instead of dealing directly with every service.
- 1:10:29
Channels for Reaching Shopify Product Users
The discussion considers Facebook groups, Shopify communities, and direct outreach as possible user-acquisition channels, contrasted with the desire for more automated traffic. The practical view is that outbound may still be necessary when no sufficient automatic channel exists.
- 1:11:34
A Product Idea’s Success Depends on Audience Reach
The speaker considers Nexus a good idea with a chance of succeeding, but qualifies that judgment on finding a specific reachable audience and then developing the product further from that base.
- 1:13:31
Solving Distribution and Building an Audience First
The discussion describes traffic acquisition as one of the hardest stages of a project and argues that solving distribution before building creates a major advantage. It also offers prior audience-building as a possible explanation for some products’ easier success—an attributed opinion rather than a demonstrated conclusion.
- 1:14:45
Preventing Agents from Mixing Project Contexts
A discussion of whether keeping projects in separate folders or workspaces helps an agent retain the correct context, while acknowledging that it might link a project to the wrong service. A review agent or an additional validation layer was suggested to verify linking when this becomes a problem.
- 1:16:17
Claude Sonnet’s Speed–Token Usage Trade-off
The speaker considers Claude Sonnet faster and less expensive, while noting that it may consume tokens and usage allowances quickly. This is presented as a personal observation, not a demonstrated benchmark.
- 1:18:54
A Usage Dashboard for AI Coding Tools
A proposed “Usage” section would display token consumption and usage limits with progress bars for tools such as Codex and Claude. The goal is to make remaining capacity visible before relying on more autonomous workflows.
- 1:26:22
Gradual Email Warm-Up
The speaker discusses starting with limited outreach and increasing it gradually after confirming mailbox readiness, instead of sending at high volume immediately.
- 1:27:42
What the Earlier Command Center Helped With
The speaker says a version built in March helped him plan and build Burnish, but did not help with debugging or autonomous tasks; he recalls that its structure resembled a PRD format.
- 1:29:35
Finding Stores by Business Type
Answering a question about finding contacts in specific business categories, the speaker identifies Store Leads as a source for discovering and filtering stores, including Shopify stores.
- 1:30:22
From Individual AI Tasks to Goal-Driven Missions
From a vibe-coding perspective, the speaker argues that improving AI tools can be given a broad goal rather than a sequence of small tasks. He proposes a platform that consolidates agents and uses an orchestrator to delegate work, with automated review, testing, and security checks before merging; the idea remains exploratory.
Got a question about something in this stream?
Ask in the group