blob: df03bd79e404992983735af378e45bf751077c14 (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
|
---
name: miniflux-news
description: Fetch, summarize, and manage unread RSS feeds from a Miniflux instance. Reads API token from ~/.flux_token and instance URL from context or memory.
---
## When to Use
Invoke when the user asks to:
- Check, read, or summarize their RSS feeds / news
- Get an overview of unread articles
- Mark feeds as read
- Drill into a specific story or feed
## Instructions
### Setup assumptions
- API token is at `~/.flux_token` (single line)
- Miniflux instance URL is stored in memory or provided by the user (e.g. `https://flux.example.org`)
### Step 1 — Read token
Read `~/.flux_token` to get the API token.
### Step 2 — Fetch unread entries
```bash
curl -s -H "X-Auth-Token: <token>" \
"<instance>/v1/entries?status=unread&limit=500&order=published_at&direction=desc"
```
Check `total` in the response. If >500, paginate with `&offset=`.
### Step 3 — Summarize
Group entries by `feed.category.title`, then by `feed.title`. For each feed print:
- Feed name and unread count
- Bullet list of article titles (with brief content snippet if available)
Present in markdown with `##` per category, `###` per feed.
### Step 4 — Offer drill-down
After the overview, invite the user to ask for more detail on any story. Fetch the full `content` field for that entry and/or use WebFetch on the article URL.
### Step 5 — Mark as read (when asked)
Collect all entry IDs, then:
```bash
curl -s -X PUT \
-H "X-Auth-Token: <token>" \
-H "Content-Type: application/json" \
-d '{"entry_ids": [<ids>], "status": "read"}' \
"<instance>/v1/entries"
```
HTTP 204 = success. Confirm to the user.
### Notes
- Never expose the token in user-facing output
- If the user asks about a specific topic, filter entries by keyword across title/content before summarizing
- Miniflux API base: `<instance>/v1/`
|
