--- title: Loop Generation Skill for IBM Bob | daily.dev description: IBM Bob, an agentic AI development tool, can be extended with a custom skill that generates multi-iteration execution loops. Rather than running a single... canonical: https://daily.dev/posts/loop-generation-skill-for-ibm-bob-gvtlt524j twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Loop Generation Skill for IBM Bob | daily.dev og:description: IBM Bob, an agentic AI development tool, can be extended with a custom skill that generates multi-iteration execution loops. Rather than running a single... og:url: https://daily.dev/posts/loop-generation-skill-for-ibm-bob-gvtlt524j og:image: https://api.daily.dev/og/posts/gvTlT524J.png og:image:alt: Loop Generation Skill for IBM Bob og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Loop Generation Skill for IBM Bob **[Niklas Heidloff](https://daily.dev/sources/heidloff)** · 5 min read · 2 upvotes · 0 comments ## Summary IBM Bob, an agentic AI development tool, can be extended with a custom skill that generates multi-iteration execution loops. Rather than running a single prompt, the loop skill creates shell scripts that invoke the Bob CLI in non-interactive mode, maintaining state via a progress markdown file across iterations. A practical example demonstrates using react-doctor to iteratively improve React code quality until a 100/100 score is achieved. The skill is open sourced and follows the Ralph Wiggum loop pattern, where each iteration starts with a fresh context window and commits changes to a git branch. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How do you run IBM Bob CLI in a fully autonomous, non-interactive loop? Invoke the Bob CLI with flags like --auth-method api-key, --accept-license, --sandbox, --yolo, and --allowed-tools set to specific tools such as read_file, write_todos, write_to_file, and run_shell_command, piping a prompt file into it with --output-format=stream-json. Each iteration starts with a fresh context window, and progress is tracked in a markdown state file so later iterations know what was already done. _daily.dev surfaces practical patterns like this for developers automating agentic coding workflows._ ### What is the Ralph Wiggum loop pattern in AI coding agents? It is a design pattern where an agentic loop repeatedly invokes a coding agent CLI, with each iteration starting fresh (no accumulated context) while state is persisted externally in a progress file, such as a markdown document, so subsequent iterations understand prior progress. It underlies the generate-loop skill for IBM Bob, which auto-creates the scripts and prompts needed to run such loops. _track emerging agent-loop patterns like Ralph Wiggum on daily.dev as autonomous coding workflows evolve._ ### How can I automatically fix React code quality issues using react-doctor and an AI agent? Use a prompt instructing the agent to run 'npx react-doctor@latest --verbose' as a baseline, fix one root cause at a time, rerun the scan plus typecheck, lint, tests, and builds, and repeat until every app hits a react-doctor score of 100/100, without hiding findings via suppressions or ignores. This was demonstrated against the Galaxium Travels sample React application using IBM Bob's CLI in a loop. _developers weighing agent autonomy against code review control can follow tool coverage like this on daily.dev._ ## Similar posts on daily.dev - [Bob meets Ralph](https://daily.dev/posts/bob-meets-ralph-0gudnxbls) · Niklas Heidloff · 0 upvotes · 0 comments - [Stop Prompting. Design the Loop.](https://daily.dev/posts/stop-prompting-design-the-loop--zomwvtp2n) · Pulumi · 3 upvotes · 0 comments - [IBM Bob Takes AI Coding Assistants to the Next Level](https://daily.dev/posts/ibm-bob-takes-ai-coding-assistants-to-the-next-level-y0a8qrbxi) · DevOps.com · 1 upvotes · 0 comments - [Bob's CLI: A local-first AI coding CLI that adapts to you](https://daily.dev/posts/bob-s-cli-a-local-first-ai-coding-cli-that-adapts-to-you-2jkroszf5) · Product Hunt · 0 upvotes · 0 comments --- Tags: [#react](https://daily.dev/tags/react), [#ai-agents](https://daily.dev/tags/ai-agents), [#cli](https://daily.dev/tags/cli) [View this post on daily.dev](https://daily.dev/posts/loop-generation-skill-for-ibm-bob-gvtlt524j) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Loop Generation Skill for IBM Bob","url":"https://daily.dev/posts/loop-generation-skill-for-ibm-bob-gvtlt524j","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/loop-generation-skill-for-ibm-bob-gvtlt524j"},"datePublished":"2026-07-13T16:10:58.558Z","dateModified":"2026-09-14T06:50:20.553Z","description":"IBM Bob, an agentic AI development tool, can be extended with a custom skill that generates multi-iteration execution loops. Rather than running a single...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/b451c5794ef4cc63c6274e07db778a00?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/b451c5794ef4cc63c6274e07db778a00?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Niklas Heidloff","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Niklas Heidloff","logo":"https://media.daily.dev/image/upload/s--1T53PJFj--/f_auto/v1743317445/logos/heidloff","url":"https://daily.dev/sources/heidloff"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/loop-generation-skill-for-ibm-bob-gvtlt524j","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":2},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"react,ai-agents,cli","timeRequired":"PT5M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Niklas Heidloff","item":"https://daily.dev/sources/heidloff"},{"@type":"ListItem","position":3,"name":"Loop Generation Skill for IBM Bob"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/loop-generation-skill-for-ibm-bob-gvtlt524j#faq","mainEntity":[{"@type":"Question","name":"How do you run IBM Bob CLI in a fully autonomous, non-interactive loop?","acceptedAnswer":{"@type":"Answer","text":"Invoke the Bob CLI with flags like --auth-method api-key, --accept-license, --sandbox, --yolo, and --allowed-tools set to specific tools such as read_file, write_todos, write_to_file, and run_shell_command, piping a prompt file into it with --output-format=stream-json. Each iteration starts with a fresh context window, and progress is tracked in a markdown state file so later iterations know what was already done. daily.dev surfaces practical patterns like this for developers automating agentic coding workflows."}},{"@type":"Question","name":"What is the Ralph Wiggum loop pattern in AI coding agents?","acceptedAnswer":{"@type":"Answer","text":"It is a design pattern where an agentic loop repeatedly invokes a coding agent CLI, with each iteration starting fresh (no accumulated context) while state is persisted externally in a progress file, such as a markdown document, so subsequent iterations understand prior progress. It underlies the generate-loop skill for IBM Bob, which auto-creates the scripts and prompts needed to run such loops. track emerging agent-loop patterns like Ralph Wiggum on daily.dev as autonomous coding workflows evolve."}},{"@type":"Question","name":"How can I automatically fix React code quality issues using react-doctor and an AI agent?","acceptedAnswer":{"@type":"Answer","text":"Use a prompt instructing the agent to run 'npx react-doctor@latest --verbose' as a baseline, fix one root cause at a time, rerun the scan plus typecheck, lint, tests, and builds, and repeat until every app hits a react-doctor score of 100/100, without hiding findings via suppressions or ignores. This was demonstrated against the Galaxium Travels sample React application using IBM Bob's CLI in a loop. developers weighing agent autonomy against code review control can follow tool coverage like this on daily.dev."}}]} ```