ShellTalk brings deterministic text-to-bash
Details
- External ID
- 47865778
- Source
- HN
- Company
- —
- Product
- ShellTalk brings deterministic text-to-bash
- Website domain
- barrasso.me
- Launched
- April 22, 2026
- Cohort
- —
- Upvotes
- 9
- Upvotes percentile
- 0.5501285347043702
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:26 p.m.
- Updated at
- Sept. 7, 2026, 9:26 p.m.
Description
Hi HN! I built a CLI tool called ShellTalk for macOS, Linux, and web (WebAssembly) that maps English text to the corresponding Bash commands.ShellTalk is written in Swift and available under the Apache 2.0 license on GitHub. I was inspired a few weeks ago after reading the Meta-Harness paper and seeing a tool called Hunch that did something similar using the Apple Foundation model. I often forget flag names and orders, but I wanted something that worked consistently. The 3B AFM worked surprisingly well with Hunch, but it felt slow and sometimes slight changes in what I wrote would result in very different outputs.ShellTalk attempts to match the input with an intent category (Git, File I/O, etc), then a template, and finally to slot-fill and adapt to the specific command version and BSD vs GNU syntax. It has a few other tricks including using NSSpellChecker on macOS to auto-correct certain typos, and scores the output on safety (i.e. is the action destructive or non-reversible).It's clearly far from perfect, but has very tight testing and validation cycles compared to using an LLM, is very portable, and might eventually work in other languages or environments like Windows. I'm curious to hear what others think.
Enrichment
- Theme
- ai voice dictation and transcription tools
- Vertical
- Horizontal
- Function
- Dev tools
- Audience
- Developer
- AI stance
- AI-native
- Project type
- Commercial product
- Normalized one-liner
- text to bash commands
- Manually corrected
- False
Could you build this?
Partial The author explicitly demonstrated that a working prototype was rapidly iterated with Claude Code, but the specialized pipeline requires hand-crafted semantic templates, POS tagging, and BM25/TF-IDF scoring without an LLM backend.
What it would actually take: The architecture relies on a deterministic NLP pipeline implemented in Swift: regex/POS entity extractors, BM25/TF-IDF retrieval indices, and cosine similarity over local embeddings (e.g., Apple NLEmbedding). The hard parts are maintaining extensive curated Bash command grammars across macOS BSD and Linux GNU variants and tuning deterministic ranking heuristics to achieve high accuracy without hallucinatory regressions. Requires domain knowledge of Unix toolchains and traditional deterministic information retrieval.
Discussion
No comments on this launch.
Competitors
Other products that read as similar to this one — 116 launches clear the similarity bar, closest 8 shown.
Attention rank: #63 of 117 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 174 days after the earliest competitor.
- Shifu · hn · 2026-03-30 · 7 upvotes · similarity 0.47
- Kosh · hn · 2026-08-28 · 7 upvotes · similarity 0.46
- Pianoterm · hn · 2026-03-02 · 61 upvotes · similarity 0.45
- Grsh · hn · 2026-01-14 · 50 upvotes · similarity 0.42
- TalkiTo · hn · 2025-11-14 · 5 upvotes · similarity 0.42
- Bash4LLM+ · hn · 2026-06-28 · 60 upvotes · similarity 0.42
- Pu.sh · hn · 2026-04-30 · 92 upvotes · similarity 0.42
- CLI-use · hn · 2026-04-21 · 7 upvotes · similarity 0.41
Other launches for this product
- No other launches for this product.
Same idea, different domain
Nobody's really built a dev tools tool for Sales yet.