How do I install the Hn skill?

Run npx skillmds@latest add voska/hn in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Hn?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

voska (@voska) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Hn by voska — Agent Skill

Name: Hn
Author: voska

Hacker News CLI for searching stories, reading posts with comments, browsing the front page, and looking up users. Use when you need HN context: tech discussions, Show HN projects, Ask HN threads, startup launches, developer opinions, or trending topics. Free API, no auth required.

voska 24f2d27 2.4 KB Updated Sep 17, 2026

File contents

hn -- Hacker News CLI

CLI for searching and reading Hacker News via the Algolia API. Returns compact plaintext by default (token-efficient). No authentication required.

Quick Reference

hn search "query"                    # Search stories (default)
hn search "query" --comments         # Search comments instead
hn search "query" --sort date        # Sort by date (default: relevance)
hn search "query" -n 10              # Limit results
hn search "query" --after 2026-01-01 # Only results after date
hn search "query" --min-points 100   # Minimum points filter

hn front                             # Current front page stories
hn front -n 10                       # Fewer results

hn read ID                           # Read story + top comments
hn read ID --expand                  # All comments expanded

hn user USERNAME                     # User profile and karma

hn status                            # API health check
hn version                           # Version info

Aliases: hn s = hn search, hn f = hn front, hn r = hn read, hn u = hn user.

Workflow

Search or browse front page: hn search "topic" -n 5 or hn front
Note the item ID from the HN URL in results
Read the full story + comments: hn read ID
Expand all comments if needed: hn read ID --expand

Output Format

Compact plaintext by default (token-efficient). Use --json only when writing to files -- never use --json in LLM context.

Search output

1532 results:
1. [523pts 89c] Show HN: I built a tool for X  @username  2h ago
   https://example.com/article
   HN: https://news.ycombinator.com/item?id=12345

Read output

Full story title, metadata, URL, body text, and threaded comments with author and timestamps. Top 3 levels by default, --expand for all.

User output

@pg  karma:157316
Bug fixer.
https://news.ycombinator.com/user?id=pg

Exit Codes

0 success
1 general error
3 empty result (no matches)
5 not found (invalid ID or username)
8 retryable error (network/timeout)

Installation

go install github.com/voska/hn-cli/cmd/hn@latest

---name: hndescription: Hacker News CLI for searching stories, reading posts with comments, browsing the front page, and looking up users. Use when you need HN context: tech discussions, Show HN projects, Ask HN threads, startup launches, developer opinions, or trending topics. Free API, no auth required.---# hn -- Hacker News CLICLI for searching and reading Hacker News via the Algolia API. Returns compactplaintext by default (token-efficient). No authentication required.## Quick Reference```bashhn search "query" # Search stories (default)hn search "query" --comments # Search comments insteadhn search "query" --sort date # Sort by date (default: relevance)hn search "query" -n 10 # Limit resultshn search "query" --after 2026-01-01 # Only results after datehn search "query" --min-points 100 # Minimum points filterhn front # Current front page storieshn front -n 10 # Fewer resultshn read ID # Read story + top commentshn read ID --expand # All comments expandedhn user USERNAME # User profile and karmahn status # API health checkhn version # Version info```Aliases: `hn s` = `hn search`, `hn f` = `hn front`, `hn r` = `hn read`, `hn u` = `hn user`.## Workflow1. Search or browse front page: `hn search "topic" -n 5` or `hn front`2. Note the item ID from the HN URL in results3. Read the full story + comments: `hn read ID`4. Expand all comments if needed: `hn read ID --expand`## Output FormatCompact plaintext by default (token-efficient). Use `--json` only when writingto files -- never use `--json` in LLM context.### Search output```1532 results:1. [523pts 89c] Show HN: I built a tool for X @username 2h ago https://example.com/article HN: https://news.ycombinator.com/item?id=12345```### Read outputFull story title, metadata, URL, body text, and threaded comments with authorand timestamps. Top 3 levels by default, `--expand` for all.### User output```@pg karma:157316Bug fixer.https://news.ycombinator.com/user?id=pg```## Exit Codes- `0` success- `1` general error- `3` empty result (no matches)- `5` not found (invalid ID or username)- `8` retryable error (network/timeout)## Installation```bashgo install github.com/voska/hn-cli/cmd/hn@latest```

Frequently asked questions