dusk-network

dusk-network / pituitary

Public

Catch spec drift before it catches you. Indexes your specs, docs, and decision records, then detects overlap, contradictions, stale docs, and code that drifts from what was decided.

github.comdusk-networkpituitary ai-tools cli developer-tools documentation golang
14
1
100% credibility
Found Mar 27, 2026 at 14 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Go
AI Summary

Pituitary detects drifts between specifications, documentation, and code to keep software projects consistent.

How It Works

1
🔍 Discover Pituitary

You hear about a helpful friend for keeping your project's notes, guides, and plans perfectly in sync.

2
đź“Ą Get the app

Download the single ready-to-use file and place it in your project folder—no setup hassle.

3
🚀 Scan your project

Tell it to look at your decisions, guides, and notes; it quickly maps everything out.

4
🧠 Spot mismatches

It highlights where guides contradict plans or old words linger, saving you hours of manual checks.

5
🔧 Fix and review

Apply simple fixes or dive into clear reports to update what's out of date.

âś… Everything aligns

Your project stays consistent as you build, with confidence your team shares the same understanding.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 14 to 14 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is pituitary?

Pituitary indexes your specs, docs, and decision records into a local SQLite database, then scans for overlaps, contradictions, stale references, and code changes that drift from agreed decisions. Pipe a git diff into `pituitary check-compliance` before merging PRs, or run `check-doc-drift --scope all` to flag outdated guides—catches spec drift like handling a catch-22 on GitHub or specific exceptions in Python/C#. Built in Go as a single binary, it runs offline-first with optional semantic embeddings via OpenAI-compatible APIs.

Why is it gaining traction?

Zero-setup CLI hooks into CI (`pituitary index --rebuild` then compliance checks), auto-fixes deterministic drift with `fix --dry-run`, and exposes an MCP server for AI agents in editors like Cursor—search specs semantically or review changes mid-session. Unlike grep or manual audits, it traces impact across artifacts and suggests `supersedes` relations, preventing silent contradictions as teams scale docs.

Who should use this?

Backend teams enforcing specs in CI/CD pipelines, where code must match ADRs. Docs-heavy orgs fighting stale runbooks after spec evolutions. Devs using AI for spec writing, needing to catch terminology drift like "repo" to "locality" across guides.

Verdict

Early-stage with 14 stars and 1.0% credibility—active development, solid README/cheatsheet, but light on users. Worth dogfooding if spec hygiene bites you; rebuilds are fast, docs guide CI integration well.

(187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.