AI Decision System
Replayable, offline-verifiable records for AI-assisted go/no-go decisions: experiment promotions, typed model answers and human-approved actions.
Evidence-first, policy-constrained developer storage autopilot for macOS.
Translation pending
Developer toolsRust · SwiftGitHub @Chisanan232OnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
Owners of public repositories under a personal GitHub account can sign in to verify ownership and manage the introduction themselves. No administrator approval is needed.
Sign in, generate a verification file, and commit it to this repository’s default branch within 30 minutes. Files on other branches or forks do not count. Then check the file below to manage the project without administrator approval. Keep the file to retain your access.
If needed, you will go to sign-in. After signing in, return here and press your chosen verification button again.
Run /nomorevibe verify in your project folder to verify ownership and manage the introduction yourself.
Verify in one minuteAsk about other verification options
Evidence-first, policy-constrained developer storage autopilot for macOS. Recover disk pressure safely: AI can recommend, policy decides, executor verifies, filesystem reality wins.
Translation pending
Taken from the repository README
Loading the full README. The saved excerpt is shown below.
Evidence-first, policy-constrained developer storage autopilot for macOS.
When a developer machine enters disk pressure, Glomeris discovers meaningful
reclaimable storage, explains why a resource is or is not safe to remove,
avoids disrupting active work, and executes only policy-approved typed
cleanup actions — re-measuring actual freed bytes until a target is reached
or no safe action remains.
AI can recommend. Policy decides. Executor verifies. Filesystem reality
wins.Read the full README ↗Evidence-first, policy-constrained developer storage autopilot for macOS.
When a developer machine enters disk pressure, Glomeris discovers meaningful reclaimable storage, explains why a resource is or is not safe to remove, avoids disrupting active work, and executes only policy-approved typed cleanup actions — re-measuring actual freed bytes until a target is reached or no safe action remains.
AI can recommend. Policy decides. Executor verifies. Filesystem reality wins.
Glomeris is not a generic du, not a cache-path list, not a system
optimizer, and not an LLM that emits rm -rf. An optional BYOK LLM planner
may rank and explain ambiguous candidates, but it can only select from a
typed, closed set of known actions — it can never invent or execute a raw
command, and it can never upgrade a PROTECTED resource to safe.
This is an experimental MVP, not a stable release. It does not promise universal safe deletion. Read the safety model before running any non-dry-run cleanup.
Build from source (requires Rust/Cargo):
git clone https://github.com/Chisanan232/glomeris.git
cd glomeris
cargo build --release
Prebuilt macOS binaries for aarch64-apple-darwin and x86_64-apple-darwin
are published with checksums under
Releases.
There is also a menu-bar app (macos/GlomerisMenuBar), built from this repo
with Xcode. It is a thin client over the same binary — see the
Menu Bar App page.
glomeris --help # every command, grouped by what it can do
glomeris detect # what tool caches exist here
glomeris clean --dry-run # the real plan, executing nothing
detect, scan, explain, clean and llm-plan never mutate anything.
execute, free, emergency and autopilot run delete — read the
safety model first.
Full documentation is published at chisanan232.github.io/glomeris: safety model, evidence model, emergency mode, policy-constrained Autopilot, BYOK LLM configuration, the menu-bar app, architecture, and a specific accounting of known limitations.
Apache-2.0 — see LICENSE.