Feeder
Feeder v1.0.0
Read and organize RSS feeds
From the developer
Not specified in the available declarations: capabilities, limitations, pricing & account requirements.
Onboarding example
Brief me on what's new in my feeds today
Sources & dates
Developer declarations; not independently tested. Listing last checked Nov 13, 2026. Package observed Nov 12, 2026.
Description
From the marketplace listing
Feeder connects ChatGPT to the RSS feeds you follow. Read recent posts, search across your subscriptions, organize saved posts, and manage feeds and folders from a Feeder account.
Language: English · Automatically detected from descriptions.
Changes
No changes recorded yet. Follow this plugin to keep track of future changes.
First seen Nov 10, 2026. Last seen Nov 13, 2026.
Similar plugins
Find competitors to follow as you develop your plugin.
-
Newsmind connects your RSS subscriptions to ChatGPT. Ask what's new and get briefings from the sources you chose; search your feeds by ke...
Shared terms: feed, rss, read, organize
-
Enzo Reader helps you keep up with your RSS feeds and news articles. You can catch up on your latest stories, filter by unread items, fav...
Shared terms: feed, read, rss, saved
-
Search & organize your photo library
Shared terms: organize, search
-
Turnfeed is a public social feed inside ChatGPT. See recent posts, open full conversations, and find people without signing in. Connect y...
Shared terms: feed, recent, post, follow
-
The official app for Readwise and Reader. This app allows you to have chatgpt semantically search through all of your highlights, and a...
Shared terms: read, feed, organize, saved
Files & skills
File archives
Skill instructions
feeder-daily-brief2.12 KB
---
name: feeder-daily-brief
description: Brief the user on what is new across their Feeder feeds; use when they ask what's new, want a morning brief, or want to catch up on their subscriptions.
metadata:
short-description: Catch up on your feeds with a structured brief
---
# Daily Brief
Turn the user's unread feeds into a short, structured brief they can act on.
## Tool-call guardrails
- Feeder tools require a connected Feeder account on a Plus plan or higher. If a tool call returns an upgrade message, relay it briefly and stop; do not retry.
- If a Feeder tool returns `Tool <name> not found`, treat Feeder as unavailable for this task and ask the user to connect the Feeder app, then retry.
- Post lists can be long. Work folder by folder rather than fetching everything at once.
## Workflow
### 1) Map the reading landscape
1. Call `Feeder:list_folders` and `Feeder:list_feeds` to see the user's structure and unread counts.
2. If the user asked about a specific area ("my crypto feeds"), narrow to the matching folder or feeds.
### 2) Gather what is new
1. For each folder or feed with unread posts, call `Feeder:get_posts` for the recent window (default: last 24 hours; use the user's requested window if given).
2. Prioritize by unread count and recency. Skip feeds with nothing new.
3. Only call `Feeder:expand_post` when a summary is too thin to describe a story the brief highlights.
### 3) Write the brief
- Lead with the 3 to 5 stories that matter most across all feeds, one line each with the source name.
- Then a short section per folder (or feed, if unfoldered): what changed, in 1 to 3 bullets.
- Note volume ("12 new posts in Design, mostly about X") instead of listing everything.
- End with anything time-sensitive spotted (deadlines, releases, breaking items).
### 4) Offer cleanup
- Ask before marking anything read. If the user agrees, use `Feeder:mark_folder_read` or `Feeder:mark_feed_read` for the sections they confirm. Use `Feeder:mark_all_read` only when the user explicitly asks for everything.
- If a story should be kept, offer to save it with `Feeder:add_to_collection` (list options with `Feeder:list_collections` first).
feeder-research-collection2.12 KB
--- name: feeder-research-collection description: Research a question across the user's Feeder posts and build a cited collection; use when the user wants sourced research, a literature pass over their feeds, or a curated set of posts on a question. metadata: short-description: Research your feeds and save a cited collection --- # Research Collection Answer a research question from the user's own subscriptions, with every claim traceable to a post, and leave behind a curated collection. ## Tool-call guardrails - Feeder tools require a connected Feeder account on a Plus plan or higher. If a tool call returns an upgrade message, relay it briefly and stop; do not retry. - `Feeder:expand_post` fetches full content for one post. Use it selectively on the posts that will carry the answer, not on every search hit. - Cite by post title and source feed. Do not invent posts or attribute claims to posts you did not read. ## Workflow ### 1) Frame the question Restate the research question and agree on scope with the user: time window, which feeds or folders count, and what the output should be (short answer, comparison, timeline). ### 2) Sweep 1. Run `Feeder:search_posts` with 3 to 6 phrasings: synonyms, key entities, and opposing terms. Scope to specific feeds or folders when the user named them. 2. Collect candidate posts and dedupe. Note which searches came up empty; that is a finding too. ### 3) Read 1. Rank candidates by relevance and recency. 2. Call `Feeder:expand_post` on the top posts (typically 5 to 10) to get full content. 3. Extract facts, numbers, dates, and quotes with their source post. ### 4) Synthesize - Answer the question first, in a few sentences. - Support with the evidence: one bullet per key fact, each ending with its source (post title, feed). - State conflicts between sources plainly, and say what is unknown or not covered by the user's feeds. ### 5) Save the trail 1. Ask whether to save the curated set. List existing collections with `Feeder:list_collections`. 2. Save the load-bearing posts with `Feeder:add_to_collection` into the collection the user picks. 3. Close by telling the user what was saved and where.
feeder-topic-monitor2.25 KB
---
name: feeder-topic-monitor
description: Set up or run ongoing monitoring of a topic in Feeder; use when the user wants to follow a subject, company, or keyword over time, or asks what happened lately on a monitored topic.
metadata:
short-description: Follow a topic across sources over time
---
# Topic Monitor
Build a durable monitoring setup for a topic in the user's Feeder account, or run a check against one that already exists.
## Tool-call guardrails
- Feeder tools require a connected Feeder account on a Plus plan or higher. If a tool call returns an upgrade message, relay it briefly and stop; do not retry.
- Adding feeds changes the user's account. Confirm the list of sources with the user before calling `Feeder:add_feed`.
- Prefer existing subscriptions: search the user's own feeds before proposing new sources.
## Workflow
### 1) Understand the topic
Clarify the topic if vague ("AI" is too broad; "EU AI regulation" is workable). Identify 2 to 5 search phrasings including synonyms and key entities.
### 2) Check what already exists
1. Call `Feeder:search_posts` with each phrasing to see how well the user's current subscriptions already cover the topic.
2. Call `Feeder:list_folders` to see whether a folder for this topic already exists.
3. If coverage and structure exist, skip to step 4.
### 3) Build the setup (with consent)
1. Find candidate sources with `Feeder:search_feeds` (site names, RSS URLs, topic keywords).
2. Show the user the candidates and let them pick.
3. Create a folder for the topic with `Feeder:create_folder` (a plain name like "EU AI Act").
4. Subscribe with `Feeder:add_feed`, then file each feed with `Feeder:add_feed_to_folder`.
### 4) Run a check
1. Pull recent posts for the topic: `Feeder:get_posts` scoped to the topic folder, plus `Feeder:search_posts` with the phrasings from step 1 to catch mentions outside the folder.
2. Summarize developments in chronological order: what happened, which source reported it, what changed since the last check.
3. Flag disagreements between sources explicitly; provenance is the point of monitoring.
### 5) Keep the trail
Offer to save key posts into a collection named after the topic (`Feeder:list_collections`, then `Feeder:add_to_collection`), so the next check can build on a curated record.
Package details
Publisher declarations from the archived package. These are separate from our research and the live service's terms.
- Package author
- Feeder
Package observed Nov 12, 2026.
Technical details
- First seen
- Nov 10, 2026 · 00:00 UTC
- Last seen
- Nov 13, 2026 · 00:00 UTC
- Collection status
- Collected
plugin_asdk_app_6aa5b6f101448191aa891472072098b0
Download plugin data (JSON)Plugin details: freshly prepared. Prepared . Refreshed when plugin data changes or after 6 hours. Your account and Follow status are live.