Self-Hosted Writing Tools
As of August 2026, AIDiveForge tracks 8 self-hosted writing tools. The top three by verified-data score are hn-ai-summarizer, Humanize, and readability-read-aloud-web-pdf-ai-summary. Curated self-hosted writing tools tracked by AIDiveForge. Listings are verified against each tool's live website and re-checked regularly.
Last updated July 29, 2026 · 8 tools
Ranked by AIDiveForge's verified-data score: data completeness, verification recency, community rating, and real visitor engagement. How we rank · No tool can pay for placement.
1. hn-ai-summarizer
The tool runs on a schedule, pulls the top Hacker News stories, summarizes them with an AI model, and delivers a filtered, translated briefing to a Telegram channel or chat. It lives entirely on your own hardware via Docker, so no data leaves your infrastructure. That self-hosted constraint is the whole point — and also the ceiling. There is no hosted version, no web UI for non-technical users, and no notification channel beyond Telegram. If your team communicates in Slack or needs a browser-based dashboard, this does not cover that.
FreeOpen SourceSelf-hostedVerified Jul 21, 2026
2. Humanize
The two skills — `humanize` and `ai-check` — work inside Claude Code, ChatGPT, Gemini, Codex CLI, Cursor, and comparable agents, not as a hosted API but as instruction files dropped into your agent's context. `humanize` rewrites text across nine documented levers drawn from 50+ peer-reviewed sources through April 2026. `ai-check` runs the reverse: forensic scoring with quoted evidence flagging the specific phrases that read as AI-generated. Because the skills are static files, there is no server, no rate limit, and no external dependency — but there is also no adaptive learning, no feedback loop, and no guarantee a detector updated after April 2026 won't develop new signals the rules don't cover yet.
FreeOpen SourceSelf-hostedVerified Jul 8, 2026
3. readability-read-aloud-web-pdf-ai-summary
The extension processes web articles and PDFs through a local readability pipeline, strips clutter, highlights sentences as they're read aloud, and stores documents in an on-device library with semantic search. No external API calls means no data leaves the browser. The sentence-level TTS highlighting keeps you oriented in long-form content where cloud read-aloud tools often lose sync. The library persists processed documents and lets you export paginated PDFs. Where it breaks: the on-device model footprint limits summarization quality compared to hosted LLMs, and users needing cross-device sync or team-shared libraries find nothing here for them.
FreeOpen SourceSelf-hostedVerified Jul 28, 2026
4. Slopsift
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.
FreeOpen SourceSelf-hostedVerified Jul 20, 2026
5. WriteWP – AI Autoblogging For WordPress
WriteWP is a WordPress plugin that pulls articles from RSS feeds or any URL, rewrites them with your own AI provider keys, and publishes on a schedule you configure once. The bring-your-own-key model means token costs land on your provider accounts at cost, with no per-article fee on top. Provider fallback is built in — if your primary model gets rate-limited, a configured backup keeps the pipeline running. The ceiling appears when you need content that goes beyond rewriting source material: original research, proprietary angles, or editorial judgment don't fit this workflow. Teams that need those capabilities run WriteWP alongside a manual editorial process, which partially restores the bottleneck the plugin was meant to remove.
PaidSelf-hostedVerified Jul 26, 2026
6. 1pager
1pager is a Claude Code skill — a scripted prompt-plus-workflow rather than a hosted app — that takes a long document, chat thread, or directory and condenses it into a bullet-first, single-page summary, then exports both a Markdown file and a DOCX. The core constraint is deliberate: least verbosity possible, with AI-tell language explicitly targeted. The workflow is a one-shot run, not an interactive loop. At the moment it only runs inside Claude Code environments, so teams without that context have no supported path to use it. One GitHub commit marks this as early-stage; expect gaps in edge-case handling.
FreeOpen SourceSelf-hostedVerified Jun 20, 2026
7. Anyword
Most AI writing tools treat content generation as a one-off task. Anyword closes the loop by predicting how your copy will perform before you publish it, using aggregated performance data across email, landing pages, ads, and social. The core appeal is quantified: the company claims a 30% lift in business outcomes by feeding conversion and engagement signals back into the model at generation time. Pricing starts at $99/month for individuals and scales to custom enterprise contracts; the private model option addresses data security concerns for large organizations. The honest limitation: you're paying for prediction sophistication, not a faster or cheaper writer—and the value hinges on whether your content workflow actually benefits from performance forecasting rather than domain expertise and testing.
PaidFree Trial · 7 days$49/moAPISelf-hostedVerified Apr 7, 2026
8. OpenBrief
The workflow is a single desktop session: import a local file or supported web link, generate a transcript (pulling existing captions when available to skip unnecessary processing), ask questions grounded in the transcript, and export a clean Markdown file. Nothing leaves your device. That privacy guarantee is the product — not a feature tier. Where it breaks: this is a one-shot summarization and Q&A tool, not an agent. It does not connect to calendars, trigger follow-up tasks, or push notes anywhere automatically. Teams that need downstream automation — routing action items into Notion, Slack, or a CRM — have to handle that export step themselves.
FreeOpen SourceAPISelf-hostedVerified Jun 9, 2026
Listings on this page are sourced and verified by the AIDiveForge data pipeline. AIDiveForge is editorially independent — no money changes hands for inclusion.