> ## Content Index > Fetch the complete content index at: https://www.implicator.ai/llms.txt > Use this file to discover other available public pages before exploring further. # Repo Radar: 5 GitHub Projects Worth Your Week - URL: https://www.implicator.ai/repo-radar-5-github-projects-worth-your-week-16/ - Published: 2026-09-04T00:34:42.000Z - Updated: 2026-09-04T00:34:42.000Z - Description: The week's GitHub momentum sat below the model layer: the rules an agent reads, the diagrams it emits, the workspace it runs in, the harness coordinating it, and the weights themselves. Heretic is the one worth an afternoon, because it puts a number on how thin safety alignment turns out to be. - Author: Marcus Schuler - Tags: Tools & Workflows Five projects climbed GitHub this week and none of them is a model. Each sits in the layer a team can own outright: the rules an agent reads before it types, the diagrams it hands back, the workspace it runs in, the harness keeping a half-dozen from colliding, and the weights underneath. 01 ### [go-modern-guidelines](https://github.com/JetBrains/go-modern-guidelines?ref=implicator.ai) JetBrains ships its GoLand team's idiom rules as an agent skill instead of IDE documentation. The plugin reads the Go version out of go.mod, then tells the agent which language features that version permits, so it writes `slices.Contains` rather than a hand-rolled loop and `errors.AsType[T]` where Go 1.26 allows. Installs into Junie, Claude Code, Codex and Cursor. ⭐ 3,088 Go Apache-2.0 Aug 31, 2026 Difficulty 1/5 **Best fit:** Go teams whose agents keep producing pre-generics code that clears review because no reviewer remembers what landed in the last four releases. **Watch out:** it needs the Go toolchain on PATH and installs a helper CLI into a local cache on first use, which locked-down build images will refuse. [ View on GitHub →](https://github.com/JetBrains/go-modern-guidelines?ref=implicator.ai) 02 ### [Archify](https://github.com/tt-a1i/archify?ref=implicator.ai) The agent writes typed JSON, not a picture. Archify validates that intermediate form against a schema, then compiles it deterministically into one self-contained HTML file covering five diagram types, dark and light themes, and PNG, SVG and WebM export. It also diffs two validated snapshots as before, delta and after, so an architecture change can be read at review time. ⭐ 45,405 JavaScript MIT Sep 2, 2026 Difficulty 2/5 **Best fit:** teams who already ask an agent to explain a service and want the answer as a file they can attach to a design review. **Watch out:** validation proves the diagram is well-formed, not that its edges exist in the code, and the project's own docs tell you to check the facts by hand. [ View on GitHub →](https://github.com/tt-a1i/archify?ref=implicator.ai) FREE WEEKDAY MORNING BRIEFING Don’t miss the next AI story that matters. The Implicator Morning Briefing filters the AI news cycle to the stories worth your attention and explains their consequences. From San Francisco, every weekday at 4:45 a.m. Pacific, 7:45 a.m. Eastern. Email address Send me tomorrow’s briefing Check your inbox for the confirmation link. About five minutes. No hype. No spam. 03 ### [Worktrunk](https://github.com/max-sixty/worktrunk?ref=implicator.ai) A Rust CLI that addresses git worktrees by branch name, so `wt switch -c -x claude feat` replaces the usual sequence of `git worktree add`, then `cd`, then starting the agent. Adds hooks on create and merge, a unique dev-server port per tree, build-cache copying between trees, and a list view carrying CI status per branch. ⭐ 6,825 Rust MIT / Apache-2.0 Sep 3, 2026 Difficulty 2/5 **Best fit:** anyone running five or more agents at once who is now losing more time to worktree bookkeeping than to the agents themselves. **Watch out:** the `wt` binary collides with Windows Terminal, so the Winget package installs it as `git-wt` and every documented command has to be retyped. [ View on GitHub →](https://github.com/max-sixty/worktrunk?ref=implicator.ai) 04 ### [Munder Difflin](https://github.com/chaitanyagiri/munder-difflin?ref=implicator.ai) An Electron app that wraps the terminal agent CLIs a developer already pays for, Claude Code, Codex, Grok, Gemini CLI and nine others, as real processes in a pseudo-terminal, then gives each one a mailbox, a markdown memory file and a desk on a 2D office floor. A supervisor agent routes work between them and escalates only when it has to. ⭐ 6,215 TypeScript MIT Sep 3, 2026 Difficulty 3/5 **Best fit:** small teams already keeping several agent sessions open who want handoffs to happen without a person copying context between windows. **Watch out:** it is pre-release at v0.4.6, it now sends anonymous usage events, and running always-on agents against subscription hourly limits sits in territory the CLI vendors have not blessed. [ View on GitHub →](https://github.com/chaitanyagiri/munder-difflin?ref=implicator.ai) 05 ### [Heretic](https://github.com/p-e-w/heretic?ref=implicator.ai) One command strips safety alignment out of an open-weights model. Heretic runs directional ablation steered by an Optuna parameter search that co-minimizes refusals and KL divergence from the original, so the modified model keeps more of its ability than hand-tuned attempts do. On Gemma 3 12B it took refusals from 97 of 100 prompts down to 3. ⭐ 30,276 Python AGPL-3.0 Sep 3, 2026 Difficulty 5/5 **Best fit:** red teams and safety researchers who need a measured figure for how fast alignment comes off a model their company is about to publish. **Watch out:** AGPL-3.0 makes derived weights awkward to ship inside a commercial product, and the command that produces a research artifact produces a decensored model for everyone else. [ View on GitHub →](https://github.com/p-e-w/heretic?ref=implicator.ai) ⭐ Repo of the Week ### Heretic Heretic earns the slot for what it measures, not for what it removes. Refusals on Gemma 3 12B fall from 97 of 100 prompts to 3, and the modified model sits 0.16 KL divergence from the original on harmless prompts, roughly a third of the next-best hand-tuned abliteration. Those two numbers describe how thin the safety layer is: a direction in the residual stream, located by a parameter search, gone in twenty to thirty minutes on an RTX 3090\. More than 5,000 models on Hugging Face now carry the tag. Treat it as an instrument rather than a product. Take an open-weights model your company already ships or plans to ship, run Heretic against it on an isolated machine, and write down the two figures: how far the refusal rate drops, and how much capability survives the drop. That pair belongs next to the model card in the risk register, because it describes what an outside party can do to those weights the day they go public. The tool's author, Philipp Emanuel Weidmann, also ships the interpretability half: pass `--plot-residuals` and it charts how harmful and harmless prompts separate layer by layer, which is the same picture a safety team would want anyway. [View Heretic on GitHub →](https://github.com/p-e-w/heretic?ref=implicator.ai) Frequently Asked Questions How were these projects selected? Current GitHub metadata, recent activity, README clarity, practical setup path, and relevance to builders working with AI systems. Are stars enough? No. Stars measure attention. Push dates, license, issues, docs, and whether the project solves a specific workflow decide usefulness. What does the difficulty score mean? It estimates how hard the project is to test or adapt, not how impressive the underlying engineering is. Which repo should readers try first? go-modern-guidelines is the cheapest test at 1/5, and Archify is the fastest way to see output. Heretic is the more strategic experiment. What should teams check before production use? License, data retention, credential access, update speed, maintainer responsiveness, and whether the repo has a realistic rollback path. AI-generated summary, reviewed by an editor. [More on our AI guidelines](https://www.implicator.ai/about/).