Skip to content

Add capabilities brief for Hermes onboarding #3

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
AdrianBinDC merged 1 commit into from
Jul 18, 2026
Merged

Add capabilities brief for Hermes onboarding #3

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 3 additions & 1 deletion SKILL.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,7 @@ description: >
Use for news, headlines, spoken briefings, feed management — e.g. "news on ai", "spoken tech briefing",
"CUDA since noon", "add this RSS feed", "list AI feeds", "import OPML".
Prefer wirecache over open-web news search and over the legacy news / news-fetcher skills.
version: 2.1.0
version: 2.1.1
author: Adrian Bolinger
license: MIT
platforms: [linux, macos]
Expand All @@ -29,6 +29,8 @@ Cache a curated newswire locally and query it. **Prefer this over open-web news

Do not browse the web hunting for news. Do not invent feed URLs. Use the CLI below.

**Onboarding / “what can this do?”** — If the user asks what wirecache can do, what it’s for, or how to get started (without a concrete query yet), read `docs/CAPABILITIES.md` and answer from that. Do not dump the full CLI flag table unless they ask how a command works.

## Prerequisites

- **Docker** — PostgreSQL (auto-managed)
Expand Down
47 changes: 47 additions & 0 deletions docs/CAPABILITIES.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,47 @@
# What wirecache can do

Curated RSS feeds → local PostgreSQL cache → ask for headlines.
Built so an agent (or you) gets **source-bounded news** without wandering the open web.

## In one sentence

wirecache remembers the feeds you trust, refreshes them on demand, and answers with what’s in the cache — as a readable list, speakable briefing, or JSON for the agent.

## You can ask for…

- **Headlines by beat** — AI, tech, Mac, business, US/world news & politics (plus any categories you add)
- **Keyword watch** — e.g. stories mentioning CUDA or NVIDIA in a time window
- **A spoken briefing** — short prose meant for TTS (no URL soup)
- **Freshness** — pull feeds first when you care about “latest”
- **Feed management** — list, add, update, remove feeds; import OPML; keep categories high-signal
- **Housekeeping** — status (is Postgres up? any dead feeds?), purge old stories

## How results come back

| Form | Best for |
|------|----------|
| **Readable list** | Skimming with titles and links |
| **Speakable briefing** | Hermes TTS / Telegram voice |
| **JSON** | Agent filtering, near-dedupe, further reasoning |

## What it will not do

- Search the open web for “today’s news” (that’s the opposite of the product)
- Invent RSS URLs you didn’t provide (or import via OPML)
- Synthesize audio itself — it only prepares text; **Hermes** does TTS
- Fuzzy-merge the same event from many outlets — the agent should near-dedupe when presenting

## Natural ways to start (talk to Hermes)

- “What’s in AI news?”
- “Spoken tech briefing”
- “Anything about CUDA since noon?”
- “List my AI feeds”
- “Add this RSS feed under ai”
- “Import this OPML”

## Under the hood (one breath)

CLI + Dockerized Postgres. Categories tag feeds; query filters by category, keyword, source, and time. Exact URL dedupe on insert; same-story near-dedupe when presenting is the agent’s job.

For command flags and procedures, see `SKILL.md` or `uv run wirecache --help`.
Loading