Get This Tool
Slopsift
Pricing
- Model
- Free
Summary
Most AI-writing linters are word blocklists dressed up as NLP — flag 'delve', call it done, ship the same false negative on every passive-voice hedge that slips through. SlopSift parses grammatical structure instead, catching the sentence-level construction that signals canned arguments, borrowed certainty, and accountability-dodging passives.
SlopSift builds a dependency graph per sentence using a 16 MiB quantized ONNX model that runs in Node or browser WebAssembly — no API call, no text upload, no remote judge. Rules inspect that graph for structural tells: the canned three-paragraph outline, the actorless passive hiding who made the decision, the certainty claim with no source. Output matches ESLint's JSON shape, so CI pipelines can consume findings without a custom parser. The ceiling appears when your linting needs go beyond the rule set the team has authored — there is no config surface described for writing custom rules, and the project is early enough that the documented rule count is small.
Bottom line: Use SlopSift when you need explainable, local-first structural linting that survives a privacy review — and plan for a gap when the rule you need does not exist yet and there is no documented path to write your own.
Community Performance Report Card
No community ratings yet. Be the first to rate this tool!
Community Benchmarks Community
Sign in to submit a benchmarkNo community benchmarks yet. Be the first to share a real-world data point.
Pros
Sign in to edit- Dependency-graph parsing instead of word matching, which means the linter catches canned argument structures and vague-attribution constructions that slip through any blocklist approach.
- 16 MiB quantized ONNX model runs fully on-device in Node and browser WebAssembly, so a draft never leaves the machine — which means teams with data-handling restrictions can actually use it without a policy exception.
- ESLint-shaped JSON output, so CI pipelines and existing editor tooling can consume findings without a custom integration layer.
- Three-tier severity grading (error / warning / note) keeps the writer in the loop on close calls rather than collapsing everything to a binary flag — which means the linter surfaces candidates for review instead of making editorial decisions it cannot defend.
- Open-source with self-hosted and browser execution paths, which means no vendor dependency and no runtime cost as document volume scales.
Cons
Sign in to edit- The documented rule set covers a fixed set of structural tells. Teams whose specific style problems — say, detecting hedging patterns specific to their domain or enforcing house argument structure — fall outside that set have no documented path to author custom rules. The next move is either forking the repo or staying with a more configurable linting framework.
- No API and no hosted tier means every environment that needs linting — local machines, CI runners, agent runtimes — must install and run the Node package independently. Teams used to a single hosted endpoint for all surfaces will be managing per-environment installs.
- The project is early-stage and the rule count is small relative to the full surface of AI writing patterns. Teams needing coverage across a broad, heterogeneous document corpus will hit gaps quickly and are more likely to end up on a general-purpose grammar or style linter that offers a wider plugin ecosystem.
Community Reviews
Sign in to write a reviewNo reviews yet. Be the first to share your experience.
About
- Platforms
- Browser, Node.js, CLI
- API Available
- No
- Self-Hosted
- Yes
- Last Updated
- 2026-07-20T03:22:50.384Z
Best For
Who it's for
- Writers editing AI-assisted text
- Developers maintaining consistent style in docs
- Teams wanting local, explainable linting without cloud calls
What it does well
- Linting AI-generated drafts for structural tells
- Reviewing Markdown and code comments in a project
- Running as a skill inside coding agents to preserve voice
- CI integration via ESLint-shaped JSON output
Integrations
Discussion Community
Sign in to commentNo discussion yet. Sign in to start the conversation.
Spotted incorrect or missing data? Join our community of contributors.
Sign Up to ContributeCommunity Notes & Tips Community
Sign in to contributeBe the first to contribute. General notes, observations, gotchas, and tips from people who use this tool day-to-day.
Frequently Asked Questions
- Is Slopsift free?
- Yes — Slopsift is fully free to use. There is no paid tier.
- Is Slopsift open source?
- Yes. Slopsift is open source.
- Can I self-host Slopsift?
- Yes. Slopsift supports self-hosting on your own infrastructure.
- What platforms does Slopsift support?
- Slopsift is available on: Browser, Node.js, CLI.
Hours Saved & ROI Stories Community
Sign in to contributeBe the first to contribute. Concrete time/cost savings, with context. e.g. "Cut my code review backlog from 4h to 45m per week."
Curated lists that include this category
Word-list linters tell you ‘delve’ appeared. SlopSift tells you why the sentence is structurally slop. The tool runs a dependency parser — trained from a compact pretrained English encoder using structured distillation and 50 targeted examples — to map every token’s part of speech and grammatical relationship. Rules then inspect that graph for constructions that signal AI-pattern writing: canned outlines repeated across paragraphs, unsupported-certainty claims, vague attribution, and actorless passives. The CLI accepts a glob, lints Markdown and code comments, and emits ESLint-shaped JSON for CI. The browser demo updates on pause, no account required.
The model is the differentiating detail. The vendor states SlopSift is not an API wrapper — they trained a 16 MiB quantized ONNX model specifically for parts-of-speech tagging and dependency parsing, with separate template families held out for evaluation. That model runs on-device in both Node and browser WebAssembly, which means the draft never leaves the machine. For teams in regulated industries or with strict data-handling policies, that is not a nice-to-have; it is the only architecture that clears the review.
SlopSift fits cleanly into a writing-assist loop where a human edits AI-generated drafts and wants a linter that explains what it caught and why — not just a confidence score from a black-box detector. It also fits as a skill inside a coding agent, letting the agent run the real linter and interpret structured findings rather than guessing at style. The wall appears at rule coverage: the documented rule set covers a defined set of structural tells, and teams whose writing problems fall outside that set have no documented path to extend the rule authoring themselves.
Install runs via `bunx slopsift .` with no prior setup. The `–format json` flag produces machine-readable output; `–level info` exposes the full candidate pile including notes. Findings are severity-graded: errors are strong tells, warnings need attention, notes are candidates for writer review — keeping editorial judgment with the human rather than collapsing everything to a pass/fail.