tw93/Mole — Repo Appraisal

Overview

What it is — A macOS CLI maintenance toolkit that combines deep cache cleaning, smart app uninstallation, disk space analysis, system optimization, and live hardware monitoring in a single mo binary.
Problem — Mac users previously needed four separate paid applications (CleanMyMac, AppCleaner, DaisyDisk, iStat Menus) to perform routine system hygiene; Mole replaces all of them for free from the terminal.
Who it's for — Mac power users and developers who prefer terminal workflows and want automated system maintenance without GUI overhead or recurring paid subscriptions.
Notable — Accumulated 53,000+ stars in under 8 months — exceptional velocity for a Shell project — and ships a companion commercial Mac app (mole.fit) signalling serious long-term product intent.

Verdict

	Rating	Summary
Quality	solid (19/24)	Actively maintained with exemplary documentation; loses points only on engineering signals inherent to Shell — no typed language, no formal manifest, no visible test infrastructure.
PAI Relevance	integrate (0.63)	A subprocess-callable macOS maintenance CLI filling a clear gap — PAI has no system maintenance skill — constrained to the Mac host side of the stack.

Quality Assessment

19/24 — actively-maintained / well-documented / early-or-minimal

Health: 7/8 (actively-maintained)

Failed:

H8: FAIL — README badges cover stars, version, license, commit-activity, Twitter, and Telegram; no CI/build-passing badge and no reference to .github/workflows/ appears in the provided data.

Passed:

H1: PASS — Release V1.39.1 tagged 2026-05-23.
H2: PASS — Most recent release is 4 days before appraisal date.
H3: PASS — Last commit 2026-05-27, well within 6 months.
H4: PASS — Last commit 2026-05-27, same day as appraisal.
H5: PASS — archived: false.
H6: PASS — 4 open issues: greater than 0 and far fewer than 100.
H7: PASS — MIT license declared.

Documentation: 8/8 (well-documented)

Failed:

(none)

Passed:

D1: PASS — README is present and substantive.
D2: PASS — README is extensive, well over 1000 bytes with full feature walkthroughs.
D3: PASS — Homebrew install (brew install mole) and curl-script install with optional version pinning documented.
D4: PASS — Multiple annotated code blocks show every command with realistic sample terminal output.
D5: PASS — Comprehensive command and flag reference covering --dry-run, --debug, --whitelist, --nightly, --json across all subcommands.
D6: PASS — First paragraph explicitly states the tool's purpose and positions it against known paid alternatives.
D7: PASS — Links to SECURITY.md, SECURITY_AUDIT.md, mole.fit commercial app, and a YouTube tutorial.
D8: PASS — Security & Safety Design section explicitly documents path validation, whitelist boundaries, macOS 15 Local Network permission limitation, and Trash-vs-delete behavior differences.

Engineering Signals: 4/8 (early-or-minimal)

Failed:

E1: FAIL — Primary language is Shell, not in the typed-language list (TypeScript, Rust, Go, Java, Kotlin, C#, Swift, Scala, Haskell).
E2: FAIL — No dependency manifest (package.json, Cargo.toml, go.mod, pyproject.toml) present or available.
E3: FAIL — No manifest available to count or verify dependencies.
E4: FAIL — README contains no mention of tests, test scripts, or CI test runs.

Passed:

E5: PASS — 53,388 stars represents substantial non-trivial adoption.
E6: PASS — ~6,591 stars/month over ~8.1 months of existence; far exceeds the 2/month threshold.
E7: PASS — 1,704 forks.
E8: PASS — Description is meaningful, specific, and well over 20 characters.

PAI Relevance

Dimension	Score	Assessment
Harvest Value	1	The append-only `~/Library/Logs/mole/operations.log` with `mo history --json` and the whitelist protection system are interesting patterns for auditable, reversible CLI operations — applicable to how PAI hooks log destructive actions — but not architecturally novel enough relative to PAI's existing hook infrastructure to score 2.
Integration Readiness	1	Shell binary callable via `bun.spawnSync("mo", ["clean", "--dry-run"])` with clean exit codes and human-readable output; Homebrew-installable on Mac in one command. Cannot run on PAI's Linux VM — macOS-only — so integration is restricted to Mac-host subprocess contexts.
Overlap Risk	0	No PAI skill or hook covers Mac system maintenance, disk analysis, cache cleanup, or app uninstallation. StacksUp manages tool subscriptions but performs no filesystem cleanup or system monitoring. Confirmed clear gap.
Gap Fill	1	Addresses local Mac host maintenance — a functional area with zero PAI coverage. This is operational rather than intelligence-layer, so it does not constitute a core AI infrastructure gap (score 2), but it is a genuine functional void for a Mac-based PAI operator.

Composite: 0.63

What Next

Capture-to-Knowledge pipeline maintenance: Install Mole with the one-liner from the README (curl …/install.sh | sh), then add mo clean as the opening step in your pipeline's shell wrapper — this clears accumulated temp JSON responses, LLM API caches, and browser-session artifacts before each ingestion run, keeping disk pressure predictable across multi-day batch sessions without manual intervention.
Fabric pattern evaluation runs (fab v2.1+): Launch mo monitor in a terminal split alongside any Fabric pattern eval that processes large content batches — the live CPU, RAM, and disk I/O readout surfaces bottlenecks in real time without leaving the terminal, replacing the need to alt-tab to Activity Monitor when a pattern pipeline stalls or a Bun process spikes unexpectedly.
Pre-download disk audit for pipeline experiments: Before pulling model weights, large corpora, or new Fabric pattern sets, run mo analyze from the home directory — it produces a ranked breakdown of actual space consumers (Xcode DerivedData, Docker image layers, bun/pnpm caches) rather than a raw du dump, giving a concrete list of reclaimable gigabytes before you commit to a large pull.

Landscape Position

Category: Developer Tools & CLIs

In this category: casey--just (excellent, 21/24)

Standing: Mole (solid, 19/24) sits one tier below just in the same category but covers an entirely different problem — just targets developer task automation across any OS while Mole targets Mac system hygiene; they serve complementary, non-overlapping workflows.

Evidence Base

Density: 9/10 — README (complete, ~8KB provided), star count, fork count, open issue count, creation date, last commit date, full release history, license, and topics all present. Dependency manifest not available (Shell project with no formal manifest convention). No CI configuration visible in provided data.

Notes

The star velocity (~6,600/month) for an 8-month-old Shell project is exceptional and indicates genuine product-market fit at scale. The companion commercial app (mole.fit — "One license covers 2 Macs, with lifetime updates") signals that the author treats this as a serious, long-term supported product rather than a side project, which is a strong maintenance durability indicator. The engineering score (4/8) reflects Shell's inherent lack of typed-language tooling and formal package management, not poor implementation discipline — the README's extensive safety documentation (SECURITY.md, SECURITY_AUDIT.md, whitelist system, dry-run modes, operation log) demonstrates careful engineering practice within the Shell idiom. The prior rolling-summary score of 4.6/24 ("poor") appears to have been an informal rough estimate; probe-based scoring yields 19/24 ("solid").