Your team fixes a logout bug.
They keep sign-ins on the server so a stolen key stops working as soon as someone logs out.
Hunch learns how your project works, then activates the right DNA Strand for every task—giving AI the context, tools, and boundaries it needs before changing your code.
It sees files, not why decisions were made, what failed before, or what a small change could break.
They keep sign-ins on the server so a stolen key stops working as soon as someone logs out.
The old discussion is hard to find. The safer code now looks more complicated than it needs to be.
The change looks clean, but it brings the same logout bug back.
The agent sees why the code exists, what it affects, and where it must stop before it edits.
A small, focused slice of Project DNA guides one assignment from start to finish.
Hunch learns how the code connects, why choices were made, what failed, and what the team expects.
It selects the knowledge, working area, risks, and boundaries for this task.
The Strand gives the agent the right tools before, during, and after the change.
Hunch sends only what the task needs—not the whole project, an old chat, or a giant prompt. Less noise and fewer wasted tokens.
npm i -g @davesheffer/hunch
Small enough to keep the agent focused. Strong enough to keep the work inside the lines. See how that works →
The active Strand helps the agent understand, plan, stay in scope, and check its work.
whyShows the decisions, rejected ideas, and past failures behind the code.
blast radiusShows which files, features, and systems a change could affect.
scopeGives the agent a clear working area and keeps unrelated changes out.
compareChecks different approaches against what the project already knows.
conformMakes sure the change still follows the project’s structure and architecture.
change gateCatches conflicts with rules your team chose to protect, with the source attached.
Want the technical details? Read how Hunch works →
One command connects Hunch to your project and coding tools.
Hunch reads how your project works and connects that knowledge to the code.
Hunch activates a fresh DNA Strand whenever an agent needs one.
# install from npm — Node 22.13+ $ npm install -g @davesheffer/hunch # connect Hunch to this project $ cd your-repo && hunch init # learn from the last 90 days $ hunch backfill --since 90d # see your project's DNA $ hunch dna inspect # ask what a file is for $ hunch why src/auth/session.ts
npm i -g @davesheffer/hunch && hunch init
Hunch gives advice by default. It will not block a change unless your team turns that on.
Claude Code? Install as a plugin instead:
/plugin marketplace add davesheffer/hunch → /plugin install hunch@hunch
Open VSX
Less noise, better decisions, safer changes.