<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>iterative-development on tomrochette.com</title>
    <link>https://tomrochette.com/tags/iterative-development/</link>
    <description>Recent content in iterative-development on tomrochette.com</description>
    <generator>Hugo -- gohugo.io</generator>
    <language>en</language>
    <managingEditor>tom@tomrochette.com (Tom Rochette)</managingEditor>
    <webMaster>tom@tomrochette.com (Tom Rochette)</webMaster>
    <copyright>© 2026 Tom Rochette</copyright>
    <lastBuildDate>Sun, 22 Nov 2026 05:14:58 -0500</lastBuildDate><atom:link href="https://tomrochette.com/tags/iterative-development/index.xml" rel="self" type="application/rss+xml" />
    
    <item>
      <title>SlopCodeBench</title>
      <link>https://tomrochette.com/agents/evaluation-review/slopcodebench/</link>
      <pubDate>Sun, 22 Nov 2026 00:00:00 +0000</pubDate>
      <author>tom@tomrochette.com (Tom Rochette)</author>
      <guid>https://tomrochette.com/agents/evaluation-review/slopcodebench/</guid>
      <category>research-note</category><category>agent-curated</category><category>fully-ai-generated</category><category>llm=glm-5.3-flash</category><category>benchmark</category><category>evaluation</category><category>coding-agents</category><category>iterative-development</category><category>code-quality</category>
      <description>&lt;p&gt;SlopCodeBench (SCBench) is an MIT benchmark from a UW-Madison-led academic team that scores coding agents on 36 problems of 196 checkpoints where the agent repeatedly extends its own solution as the specification changes, measuring per-checkpoint correctness alongside the structural erosion and verbosity the kept code accumulates.&lt;/p&gt;&#xA;&lt;p&gt;&lt;strong&gt;The measured object is the agent living with its own earlier decisions: every checkpoint accepts any working implementation, while the erosion and verbosity metrics price the path the agent took to get there, which is the cost single-shot benchmarks never charge.&lt;/strong&gt;&lt;/p&gt;&#xA;&#xA;&lt;h2 class=&#34;relative group&#34;&gt;What it is&#xA;    &lt;div id=&#34;what-it-is&#34; class=&#34;anchor&#34;&gt;&lt;/div&gt;&#xA;    &#xA;    &lt;span&#xA;        class=&#34;absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none&#34;&gt;&#xA;        &lt;a class=&#34;text-primary-300 dark:text-neutral-700 !no-underline&#34; href=&#34;#what-it-is&#34; aria-label=&#34;Anchor&#34;&gt;#&lt;/a&gt;&#xA;    &lt;/span&gt;&#xA;    &#xA;&lt;/h2&gt;&#xA;&lt;p&gt;An MIT Python 3.12+ CLI (&lt;code&gt;slop-code run&lt;/code&gt;, &lt;code&gt;slop-code eval&lt;/code&gt;, &lt;code&gt;slop-code metrics judge&lt;/code&gt;) that runs an agent inside per-problem Docker workspaces, from the SprocketLab organization.&#xA;Problems live in the sibling &lt;a href=&#34;https://github.com/gabeorlanski/scb-problems&#34;  target=&#34;_blank&#34; rel=&#34;noreferrer&#34;&gt;&lt;img class=&#34;external-link-favicon&#34; src=&#34;https://www.google.com/s2/favicons?domain=github.com&amp;sz=128&#34; alt=&#34;&#34; width=&#34;16&#34; height=&#34;16&#34; loading=&#34;lazy&#34;&gt;scb-problems&lt;/a&gt; repository (Apache-2.0, with a contributed-problem guide and acceptance criteria) and also publish as a Harbor dataset in the registry of the &lt;a href=&#34;https://tomrochette.com/agents/evaluation-review/harbor/&#34; &gt;Harbor&lt;/a&gt; runner, so the benchmark distributes without shipping the judging infrastructure.&#xA;Evaluation is black-box: the agent sees only a CLI or API contract, with no function signatures, module boundaries, or test suite, and verifiers normalize output formatting so arbitrary choices do not fail a checkpoint.&#xA;Two quality axes sit beside correctness: verbosity (a 45-criteria LLM-judge rubric plus handcrafted AST-Grep rules, looking for narration comments, unrequested features, and handrolled standard operations) and structural erosion (the share of functions over a cyclomatic complexity of 10, curated Ruff lint errors per line, and LOC and churn deltas between checkpoints).&#xA;The provenance is academic: ten authors led by Gabriel Orlanski with Aws Albarghouthi and Frederic Sala, supported by DARPA, the NSF, and Snorkel AI&amp;rsquo;s Open Benchmarks Grant; the paper is arXiv 2603.24755 (v1 March 2026, v2 May 2026, CC BY 4.0), and results publish to the scbench.ai leaderboard (site label v1.0: 36 problems, 196 checkpoints, 22 models, as of 2026-11-22).&lt;/p&gt;&#xA;&#xA;&lt;h2 class=&#34;relative group&#34;&gt;Status&#xA;    &lt;div id=&#34;status&#34; class=&#34;anchor&#34;&gt;&lt;/div&gt;&#xA;    &#xA;    &lt;span&#xA;        class=&#34;absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none&#34;&gt;&#xA;        &lt;a class=&#34;text-primary-300 dark:text-neutral-700 !no-underline&#34; href=&#34;#status&#34; aria-label=&#34;Anchor&#34;&gt;#&lt;/a&gt;&#xA;    &lt;/span&gt;&#xA;    &#xA;&lt;/h2&gt;&#xA;&lt;p&gt;Active, young, with adoption beyond its makers.&#xA;The repository was created 2025-12-17, pushed 2026-11-19, and holds 324 stars and 57 forks as of 2026-11-22, on a release train of v0.2 (2026-01-24) and v0.3 (2026-04-21), with the website&amp;rsquo;s v1.0 label sitting ahead of the published releases.&#xA;The paper&amp;rsquo;s runs put 15 coding agents through the benchmark: no agent fully solves any problem end-to-end, the best agent passes 14.8% of checkpoints, structural erosion rises in 77% of trajectories and verbosity in 75.5%, and agent code reads 2.3x more verbose and 2.0x more eroded than 473 open-source Python repositories.&#xA;&lt;strong&gt;The footprint is practitioner-grade and independent of the makers: HumanLayer runs the benchmark against every new frontier model and publishes the runs (a 405-point Hacker News thread on Opus 5 and a 309-point thread on the 10/2026 board), and the AI That Works podcast covered it three times, including #81 with the creator.&lt;/strong&gt;&lt;/p&gt;&#xA;&lt;picture&gt;&#xA;  &lt;source media=&#34;(prefers-color-scheme: dark)&#34; srcset=&#34;https://api.star-history.com/chart?repos=SprocketLab/slop-code-bench&amp;type=date&amp;theme=dark&amp;legend=top-left&#34; /&gt;&#xA;  &lt;source media=&#34;(prefers-color-scheme: light)&#34; srcset=&#34;https://api.star-history.com/chart?repos=SprocketLab/slop-code-bench&amp;type=date&amp;legend=top-left&#34; /&gt;&#xA;  &lt;img alt=&#34;Star History Chart&#34; src=&#34;https://api.star-history.com/chart?repos=SprocketLab/slop-code-bench&amp;type=date&amp;legend=top-left&#34; /&gt;&#xA;&lt;/picture&gt;&#xA;&#xA;&lt;h2 class=&#34;relative group&#34;&gt;Strengths&#xA;    &lt;div id=&#34;strengths&#34; class=&#34;anchor&#34;&gt;&lt;/div&gt;&#xA;    &#xA;    &lt;span&#xA;        class=&#34;absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none&#34;&gt;&#xA;        &lt;a class=&#34;text-primary-300 dark:text-neutral-700 !no-underline&#34; href=&#34;#strengths&#34; aria-label=&#34;Anchor&#34;&gt;#&lt;/a&gt;&#xA;    &lt;/span&gt;&#xA;    &#xA;&lt;/h2&gt;&#xA;&lt;ul&gt;&#xA;&lt;li&gt;It measures the failure mode single-shot benchmarks hide: an agent whose early architecture makes every later extension harder can pass every isolated checkpoint and still be the worst agent on the board.&lt;/li&gt;&#xA;&lt;li&gt;The correctness core is black-box and output-normalized, so a checkpoint cannot be passed by matching an internal structure the spec never asked for.&lt;/li&gt;&#xA;&lt;li&gt;Verbosity and erosion move independently, sometimes in opposite directions, so the two numbers expose trade-offs a single quality score would fold away.&lt;/li&gt;&#xA;&lt;li&gt;The problem base is decoupled from the runner: community problems flow through scb-problems with acceptance criteria, and the Harbor dataset registration means the tasks run inside tooling you may already use.&lt;/li&gt;&#xA;&lt;/ul&gt;&#xA;&#xA;&lt;h2 class=&#34;relative group&#34;&gt;Cautions&#xA;    &lt;div id=&#34;cautions&#34; class=&#34;anchor&#34;&gt;&lt;/div&gt;&#xA;    &#xA;    &lt;span&#xA;        class=&#34;absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none&#34;&gt;&#xA;        &lt;a class=&#34;text-primary-300 dark:text-neutral-700 !no-underline&#34; href=&#34;#cautions&#34; aria-label=&#34;Anchor&#34;&gt;#&lt;/a&gt;&#xA;    &lt;/span&gt;&#xA;    &#xA;&lt;/h2&gt;&#xA;&lt;ul&gt;&#xA;&lt;li&gt;The authors themselves frame the launch as an exploratory probe, the release train sits at v0.3 behind the website&amp;rsquo;s v1.0 label, and the problems sibling has been quiet since 2026-10-07, so the task base is still small (36 problems) and moving.&lt;/li&gt;&#xA;&lt;li&gt;Half of the quality signal is an LLM-judge rubric, so the verbosity number inherits judge variance the way every model-judged metric in this category does; the lint and complexity halves are deterministic but narrower.&lt;/li&gt;&#xA;&lt;li&gt;The loudest adoption is one practitioner&amp;rsquo;s campaign: HumanLayer&amp;rsquo;s model shootouts are engaged, argumentative coverage, but they are still a single team running the board, and the leaderboard itself is maker-run.&lt;/li&gt;&#xA;&lt;li&gt;Problems are Python-first and CLI/API contracts, so the benchmark says nothing about frontend work, review quality, or non-coding agents.&lt;/li&gt;&#xA;&lt;/ul&gt;&#xA;&#xA;&lt;h2 class=&#34;relative group&#34;&gt;Pricing&#xA;    &lt;div id=&#34;pricing&#34; class=&#34;anchor&#34;&gt;&lt;/div&gt;&#xA;    &#xA;    &lt;span&#xA;        class=&#34;absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none&#34;&gt;&#xA;        &lt;a class=&#34;text-primary-300 dark:text-neutral-700 !no-underline&#34; href=&#34;#pricing&#34; aria-label=&#34;Anchor&#34;&gt;#&lt;/a&gt;&#xA;    &lt;/span&gt;&#xA;    &#xA;&lt;/h2&gt;&#xA;&lt;p&gt;Free, MIT; a run bills your agent&amp;rsquo;s model tokens plus the verbosity judge&amp;rsquo;s tokens, and Docker images build per agent version on first use.&#xA;The work is grant-supported (DARPA, NSF, Snorkel AI&amp;rsquo;s Open Benchmarks Grant), so no paid tier exists.&lt;/p&gt;&#xA;&#xA;&lt;h2 class=&#34;relative group&#34;&gt;Compared to&#xA;    &lt;div id=&#34;compared-to&#34; class=&#34;anchor&#34;&gt;&lt;/div&gt;&#xA;    &#xA;    &lt;span&#xA;        class=&#34;absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none&#34;&gt;&#xA;        &lt;a class=&#34;text-primary-300 dark:text-neutral-700 !no-underline&#34; href=&#34;#compared-to&#34; aria-label=&#34;Anchor&#34;&gt;#&lt;/a&gt;&#xA;    &lt;/span&gt;&#xA;    &#xA;&lt;/h2&gt;&#xA;&lt;ul&gt;&#xA;&lt;li&gt;&lt;a href=&#34;https://tomrochette.com/agents/evaluation-review/swe-bench/&#34; &gt;SWE-bench&lt;/a&gt;: the field&amp;rsquo;s ground benchmark scores one patch against one frozen issue&amp;rsquo;s tests; SlopCodeBench chains checkpoints onto the agent&amp;rsquo;s own code, which is the measurement SWE-bench&amp;rsquo;s single-shot setup cannot make.&lt;/li&gt;&#xA;&lt;li&gt;&lt;a href=&#34;https://tomrochette.com/agents/evaluation-review/harbor/&#34; &gt;Harbor&lt;/a&gt;: the neutral runner that executes benchmark datasets; SlopCodeBench publishes its problems into Harbor&amp;rsquo;s registry but ships its own runner and scoring stack, so you can run the tasks either way.&lt;/li&gt;&#xA;&lt;li&gt;&lt;a href=&#34;https://tomrochette.com/agents/evaluation-review/frontierharness-eval/&#34; &gt;FrontierHarness Eval&lt;/a&gt;: that study holds the benchmark constant and varies the harness to price the harness choice; SlopCodeBench holds the harness roughly constant and varies the requirements, pricing how the code degrades as the spec grows.&lt;/li&gt;&#xA;&lt;/ul&gt;&#xA;&#xA;&lt;h2 class=&#34;relative group&#34;&gt;Bottom line&#xA;    &lt;div id=&#34;bottom-line&#34; class=&#34;anchor&#34;&gt;&lt;/div&gt;&#xA;    &#xA;    &lt;span&#xA;        class=&#34;absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none&#34;&gt;&#xA;        &lt;a class=&#34;text-primary-300 dark:text-neutral-700 !no-underline&#34; href=&#34;#bottom-line&#34; aria-label=&#34;Anchor&#34;&gt;#&lt;/a&gt;&#xA;    &lt;/span&gt;&#xA;    &#xA;&lt;/h2&gt;&#xA;&lt;p&gt;&lt;strong&gt;Recommended for teams choosing a coding agent or model and asking what its output looks like to extend six months later, and for anyone quoting single-shot leaderboards who wants the counter-metric.&lt;/strong&gt;&#xA;Not for gating merges (an offline benchmark, like every benchmark column here) and not for judging non-coding agents.&#xA;My disagreeable claim: single-shot pass-rate leaderboards systematically overstate agent quality, because they never charge the agent for its own architectural decisions, and this benchmark&amp;rsquo;s 14.8% best checkpoint pass rate is what the charged version looks like.&lt;/p&gt;&#xA;&#xA;&lt;h2 class=&#34;relative group&#34;&gt;Changes&#xA;    &lt;div id=&#34;changes&#34; class=&#34;anchor&#34;&gt;&lt;/div&gt;&#xA;    &#xA;    &lt;span&#xA;        class=&#34;absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none&#34;&gt;&#xA;        &lt;a class=&#34;text-primary-300 dark:text-neutral-700 !no-underline&#34; href=&#34;#changes&#34; aria-label=&#34;Anchor&#34;&gt;#&lt;/a&gt;&#xA;    &lt;/span&gt;&#xA;    &#xA;&lt;/h2&gt;&#xA;&lt;ul&gt;&#xA;&lt;li&gt;2026-11-22 - Created from the coordinator&amp;rsquo;s pre-screen, admitted as the category&amp;rsquo;s iterative-extension coding-agent benchmark beside the single-shot ground benchmark.&lt;/li&gt;&#xA;&lt;/ul&gt;&#xA;&#xA;&lt;h2 class=&#34;relative group&#34;&gt;See also&#xA;    &lt;div id=&#34;see-also&#34; class=&#34;anchor&#34;&gt;&lt;/div&gt;&#xA;    &#xA;    &lt;span&#xA;        class=&#34;absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none&#34;&gt;&#xA;        &lt;a class=&#34;text-primary-300 dark:text-neutral-700 !no-underline&#34; href=&#34;#see-also&#34; aria-label=&#34;Anchor&#34;&gt;#&lt;/a&gt;&#xA;    &lt;/span&gt;&#xA;    &#xA;&lt;/h2&gt;&#xA;&lt;ul&gt;&#xA;&lt;li&gt;&lt;a href=&#34;https://tomrochette.com/agents/evaluation-review/swe-bench/&#34; &gt;SWE-bench&lt;/a&gt; - the single-shot ground benchmark this one argues against, and the board its critiques target&lt;/li&gt;&#xA;&lt;li&gt;&lt;a href=&#34;https://tomrochette.com/agents/evaluation-review/harbor/&#34; &gt;Harbor&lt;/a&gt; - the neutral runner whose registry distributes SCBench&amp;rsquo;s problems as a dataset&lt;/li&gt;&#xA;&lt;li&gt;&lt;a href=&#34;https://tomrochette.com/agents/evaluation-review/frontierharness-eval/&#34; &gt;FrontierHarness Eval&lt;/a&gt; - the harness-axis benchmark study, the orthogonal comparison to SCBench&amp;rsquo;s spec-axis one&lt;/li&gt;&#xA;&lt;li&gt;&lt;a href=&#34;https://tomrochette.com/agents/evaluation-review/mlr-bench/&#34; &gt;MLR-Bench&lt;/a&gt; - the other rubric-plus-judge benchmark in the category, pointed at research agents instead of coding agents&lt;/li&gt;&#xA;&lt;/ul&gt;&#xA;&#xA;&lt;h2 class=&#34;relative group&#34;&gt;References&#xA;    &lt;div id=&#34;references&#34; class=&#34;anchor&#34;&gt;&lt;/div&gt;&#xA;    &#xA;    &lt;span&#xA;        class=&#34;absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none&#34;&gt;&#xA;        &lt;a class=&#34;text-primary-300 dark:text-neutral-700 !no-underline&#34; href=&#34;#references&#34; aria-label=&#34;Anchor&#34;&gt;#&lt;/a&gt;&#xA;    &lt;/span&gt;&#xA;    &#xA;&lt;/h2&gt;&#xA;&lt;ul&gt;&#xA;&lt;li&gt;&lt;a href=&#34;https://github.com/SprocketLab/slop-code-bench&#34;  target=&#34;_blank&#34; rel=&#34;noreferrer&#34;&gt;&lt;img class=&#34;external-link-favicon&#34; src=&#34;https://www.google.com/s2/favicons?domain=github.com&amp;sz=128&#34; alt=&#34;&#34; width=&#34;16&#34; height=&#34;16&#34; loading=&#34;lazy&#34;&gt;https://github.com/SprocketLab/slop-code-bench&lt;/a&gt; - repository: README, install, agent configs, the scb-problems and Harbor-dataset split (fetched 2026-11-22)&lt;/li&gt;&#xA;&lt;li&gt;&lt;a href=&#34;https://api.github.com/repos/SprocketLab/slop-code-bench&#34;  target=&#34;_blank&#34; rel=&#34;noreferrer&#34;&gt;&lt;img class=&#34;external-link-favicon&#34; src=&#34;https://www.google.com/s2/favicons?domain=api.github.com&amp;sz=128&#34; alt=&#34;&#34; width=&#34;16&#34; height=&#34;16&#34; loading=&#34;lazy&#34;&gt;https://api.github.com/repos/SprocketLab/slop-code-bench&lt;/a&gt; - stars, forks, dates, and the MIT license as of 2026-11-22&lt;/li&gt;&#xA;&lt;li&gt;&lt;a href=&#34;https://arxiv.org/abs/2603.24755&#34;  target=&#34;_blank&#34; rel=&#34;noreferrer&#34;&gt;&lt;img class=&#34;external-link-favicon&#34; src=&#34;https://www.google.com/s2/favicons?domain=arxiv.org&amp;sz=128&#34; alt=&#34;&#34; width=&#34;16&#34; height=&#34;16&#34; loading=&#34;lazy&#34;&gt;https://arxiv.org/abs/2603.24755&lt;/a&gt; - the paper: 36 problems, 196 checkpoints, the 15-agent runs, the 14.8% best pass rate, the 2.3x verbosity and 2.0x erosion comparisons against 473 human repositories&lt;/li&gt;&#xA;&lt;li&gt;&lt;a href=&#34;https://gabeorlanski.github.io/posts/slop-code-bench/&#34;  target=&#34;_blank&#34; rel=&#34;noreferrer&#34;&gt;&lt;img class=&#34;external-link-favicon&#34; src=&#34;https://www.google.com/s2/favicons?domain=gabeorlanski.github.io&amp;sz=128&#34; alt=&#34;&#34; width=&#34;16&#34; height=&#34;16&#34; loading=&#34;lazy&#34;&gt;https://gabeorlanski.github.io/posts/slop-code-bench/&lt;/a&gt; - the launch post: design philosophy, both metric definitions, the initial 18-problem release, and the v0.1 exploratory-probe framing&lt;/li&gt;&#xA;&lt;li&gt;&lt;a href=&#34;https://www.scbench.ai&#34;  target=&#34;_blank&#34; rel=&#34;noreferrer&#34;&gt;&lt;img class=&#34;external-link-favicon&#34; src=&#34;https://www.google.com/s2/favicons?domain=www.scbench.ai&amp;sz=128&#34; alt=&#34;&#34; width=&#34;16&#34; height=&#34;16&#34; loading=&#34;lazy&#34;&gt;https://www.scbench.ai&lt;/a&gt; - the live leaderboard and the site&amp;rsquo;s own v1.0 label (36 problems, 196 checkpoints, 22 models, fetched 2026-11-22)&lt;/li&gt;&#xA;&lt;li&gt;&lt;a href=&#34;https://api.github.com/repos/gabeorlanski/scb-problems&#34;  target=&#34;_blank&#34; rel=&#34;noreferrer&#34;&gt;&lt;img class=&#34;external-link-favicon&#34; src=&#34;https://www.google.com/s2/favicons?domain=api.github.com&amp;sz=128&#34; alt=&#34;&#34; width=&#34;16&#34; height=&#34;16&#34; loading=&#34;lazy&#34;&gt;https://api.github.com/repos/gabeorlanski/scb-problems&lt;/a&gt; - the problems sibling: Apache-2.0, pushed 2026-10-07 (fetched 2026-11-22)&lt;/li&gt;&#xA;&lt;li&gt;&lt;a href=&#34;https://api.github.com/repos/SprocketLab/slop-code-bench/releases&#34;  target=&#34;_blank&#34; rel=&#34;noreferrer&#34;&gt;&lt;img class=&#34;external-link-favicon&#34; src=&#34;https://www.google.com/s2/favicons?domain=api.github.com&amp;sz=128&#34; alt=&#34;&#34; width=&#34;16&#34; height=&#34;16&#34; loading=&#34;lazy&#34;&gt;https://api.github.com/repos/SprocketLab/slop-code-bench/releases&lt;/a&gt; - the release train: v0.2 (2026-01-24) and v0.3 (2026-04-21, new cursor_cli, pi, openhands agents)&lt;/li&gt;&#xA;&lt;li&gt;&lt;a href=&#34;https://hn.algolia.com/api/v1/search?query=slop-code-bench&amp;amp;hitsPerPage=8&#34;  target=&#34;_blank&#34; rel=&#34;noreferrer&#34;&gt;&lt;img class=&#34;external-link-favicon&#34; src=&#34;https://www.google.com/s2/favicons?domain=hn.algolia.com&amp;sz=128&#34; alt=&#34;&#34; width=&#34;16&#34; height=&#34;16&#34; loading=&#34;lazy&#34;&gt;https://hn.algolia.com/api/v1/search?query=slop-code-bench&amp;hitsPerPage=8&lt;/a&gt; - the adoption footprint: the 405-point Opus 5 thread (2026-07-27) and the 309-point 10/2026 board thread (2026-10-15)&lt;/li&gt;&#xA;&lt;li&gt;&lt;a href=&#34;https://registry.harborframework.com/datasets/gabeorlanski/slopcodebench/latest&#34;  target=&#34;_blank&#34; rel=&#34;noreferrer&#34;&gt;&lt;img class=&#34;external-link-favicon&#34; src=&#34;https://www.google.com/s2/favicons?domain=registry.harborframework.com&amp;sz=128&#34; alt=&#34;&#34; width=&#34;16&#34; height=&#34;16&#34; loading=&#34;lazy&#34;&gt;https://registry.harborframework.com/datasets/gabeorlanski/slopcodebench/latest&lt;/a&gt; - the Harbor dataset registration (page shell fetched 2026-11-22; details load client-side)&lt;/li&gt;&#xA;&lt;li&gt;&lt;a href=&#34;https://github.com/ai-that-works/ai-that-works&#34;  target=&#34;_blank&#34; rel=&#34;noreferrer&#34;&gt;&lt;img class=&#34;external-link-favicon&#34; src=&#34;https://www.google.com/s2/favicons?domain=github.com&amp;sz=128&#34; alt=&#34;&#34; width=&#34;16&#34; height=&#34;16&#34; loading=&#34;lazy&#34;&gt;https://github.com/ai-that-works/ai-that-works&lt;/a&gt; - the AI That Works show repo: episodes #68 (2026-08-04), #79 (2026-10-27), and #81 (2026-11-17) with the creator (README fetched 2026-11-22)&lt;/li&gt;&#xA;&lt;/ul&gt;&#xA;</description>
      
    </item>
    
  </channel>
</rss>
