# Repostit vs Typefully

> Typefully scores 49.3 (D) to Repostit's 32.5 (F) for social posting. Prices, MCP, x402, uptime and agent notes side by side.

- Canonical: https://www.anchorterminal.com/compare/repostit-vs-typefully
- Markdown: https://www.anchorterminal.com/compare/repostit-vs-typefully.md (~4,950 tokens)
- Site index for agents: https://www.anchorterminal.com/llms.txt (the key docs and every listing in one file, at most 16 MB, about 4 million tokens: https://www.anchorterminal.com/llms-full.txt)
- API: https://www.anchorterminal.com/api/v1/index.json
- MCP: https://www.anchorterminal.com/mcp (the whole catalogue as tools, no key)
- DNS: dig +short TXT <slug>.grade.anchorterminal.com (the grade as a DNS TXT record, for a sandbox with no HTTP where names still resolve; also top., best.<category>. and <a>.vs.<b>. under the zone, and the same under grade.letme.dev)
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)
- Updated: 2026-11-03

Typefully scores 49.3 (D) on agent readiness against Repostit's 32.5 (F), and leads in 5 of 7 scored categories. Repostit leads on security & auth. Both do social posting.

- Repostit: grade F, 32.5/100, rank #1191 of 1272. Markdown https://www.anchorterminal.com/tools/repostit.md · JSON https://www.anchorterminal.com/api/v1/tools/repostit.json
- Typefully: grade D, 49.3/100, rank #992 of 1272. Markdown https://www.anchorterminal.com/tools/typefully.md · JSON https://www.anchorterminal.com/api/v1/tools/typefully.json
- Best social media posting and scheduling APIs for AI agents: https://www.anchorterminal.com/best/social-media/index.md
- All 132 social comparisons: https://www.anchorterminal.com/compare/social-media/index.md

## Which one, for what

### Repostit (F)

Good for: Creators, small businesses and agencies who post short vertical video on one network and want an assistant in Claude or ChatGPT to copy it to others on a schedule, set up reposting workflows and confirm what went out, without writing code.

Ahead on:
- Security & auth, 44 against 38

Also in its favour:
- Free to start without a card
- No incidents deducted, where Typefully loses 3 points for them

Watch for: No tool names, input schemas, annotations, scopes or error shapes are published, and no public REST API, SDK or OpenAPI file exists

### Typefully (D)

Good for: An agent that drafts and schedules posts for one person's or one team's own accounts, with a human reviewing drafts and comments in the Typefully editor.

Ahead on:
- Schema & documentation, 83 against 27
- Agent ergonomics, 62 against 25
- Maintenance & community, 82 against 56
- Transparency & trust, 66 against 29

Watch for: Rate limits are described by response headers only. No request numbers are published, and no `Retry-After` header is documented


## Why the edge

### Schema & documentation: Typefully +56

- Repostit (27): MCP reading, since the connector is the only surface an outside agent reaches; no public REST API, developer page, OpenAPI file, SDK or package was found in the help centre, /whats-new/, the sitemap, llms.txt, GitHub, npm or the MCP registry. No tool names or input schemas are published anywhere, and no session was opened to read them (0 of 25). llms.txt exists, but it is a Rank Math index of site URLs that includes theme placeholder entries, and help articles are HTML only (5 of 10). The Claude connector page says what each step is for and when not to use it, list accounts before choosing a destination, a Brief saves a draft and neither publishes nor schedules, publishing applies to every destination in a Brief, Queued is not delivery, future-content workflows publish once active while existing-content ones import for review (8 of 20). Inputs are described in prose only, formats per destination, JPEG, PNG or WebP up to 20 MB, carousel caps of 4, 10 and 20 by network, media at a downloadable HTTPS URL and a timezone with every scheduled time (4 of 15). Plain-language example requests and per-network failure reasons written for people; no example tool call, response or error shape (5 of 15). The endpoint path carries v1.1 and the registry holds 2026.10.17 and 2026.10.23; /whats-new/ is undated and no changelog was found (5 of 15). 0+5+8+4+5+5 = 27.
- Typefully (83): Public OpenAPI 3.1.0 document for 24 operations and seven webhook events (25). `llms.txt`, the full reference as Markdown and an RFC 9727 API catalogue (10). Descriptions state purpose, the permission each call needs and how drafts move between states, with less on when not to use a call (15). Inputs are typed per platform with 18 enums, length limits and required fields, though `publish_at` mixes keywords and datetimes in one string (12). 239 examples and error responses on every operation, with no list of error codes (11). The version is in the path and API changes are dated in the product changelog, with no API changelog of its own (10).

### Agent ergonomics: Typefully +37

- Repostit (25): MCP reading. The tool count is not published. The connector pages describe seven actions (list accounts, create, schedule and publish a Brief, check delivery, create a workflow, send a connection link), which suggests a small set, but the definitions were not read, so a middle reading rather than full marks (10 of 25). No pagination, filtering or output-size controls are documented (0 of 20). Delivery checks report a result per destination with public links when the network gives one, and the help centre lists failure reasons per network, but no error codes or connector error messages are documented (5 of 20). No idempotency key and no readOnlyHint or destructiveHint seen; Briefs start as drafts, queued Briefs are locked against a second send, Manual workflows hold posts for approval and the guidance says to check before retrying (6 of 20). Requests are made in plain language with few stated requirements (an HTTPS media URL, a timezone); no SDK in any language (4 of 15). 10+0+5+6+4 = 25.
- Typefully (62): List calls take `limit` up to 50 with a default of 10, and a draft read can drop comment markers, but there is no field selection (15). Limit and offset pagination with status, tag and ordering filters on drafts (20). Errors carry a code, a message and field details with a location path (16). No idempotency key. A planned draft is a write that cannot publish, which makes retries safer for that state only, and MCP tool annotations were not read (4). Creating a draft needs only `platforms`, but there is no official SDK, only the skill's CLI script (7).

### Transparency & trust: Typefully +37

- Repostit (29): Closed service. The terms of service (effective 7 March 2025, no revision since) name Repostit LLC of Newark, Delaware, with Delaware law and arbitration, take a non-exclusive licence to content only to run the service and say fees are non-refundable; every footer says Repostit, Inc., and the terms do not mention the connector (12 of 30). The privacy policy (29 July 2026) keeps data until the account is deleted or as needed, with no periods, deletes Pinterest credentials on disconnect but says nothing of other networks' tokens, names no DPA, and does not mention the connector or its 90-day offline access; the account deletion article gives no timeline and leaves network-side access to the user (10 of 30). No deprecation policy or dated notices; the terms promise notice of significant changes and the ChatGPT page names a pending version 1.1 without a date (3 of 20). Subprocessors are 'hosting providers, analytics services', unnamed, with transfers outside the user's country and no locations; the Tawk.to chat on the contact page and the Google Cloud Run host of the connector are not named in the policy (2 of 20). 12+10+3+2 = 27.
- Typefully (66): Closed service under published terms, with the Agent Skill under MIT (15). The privacy policy says deleted data is purged within 30 days, that AI providers may not train on customer data and that data is processed in the United States, and it links a sub-processor list. No DPA was found, and the terms say only that data may be kept for a limited period (20). No deprecation policy. The v1 to v2 migration guide gives no end date for v1, `draft_id` is marked deprecated with no date, and the terms may change without notice (4). About 30 sub-processors listed by purpose on a page dated 3 March 2026, without locations for each (16).

### Maintenance & community: Typefully +26

- Repostit (56): Closed service. The latest registry version is 2026.10.23, published 23 October 2026, and the connector article was updated 22 October 2026 (30). Dated changes in the last 90 days are the two registry versions of 17 and 23 October 2026; /whats-new/ is undated and the ChatGPT app's 1.1.0 (per the 3 November listing check) carries no date, so fewer than three dated releases (10 of 20). Support is email and weekday chat, 9am to 5pm Eastern, with replies within 24 hours; no public changelog with dates, forum or issue tracker (6 of 15). In the official MCP registry as io.github.hax98/repostit, a GitHub-verified namespace of a personal account; the entry links Repostit's own pages and its address matches the vendor's connector page, but no first-hand page ties the account to the company, so partial credit (10 of 15). No packages, public code or CI to judge (0 of 10). 30+10+6+10+0 = 56.
- Typefully (82): Product changelog entry on 20 September 2026 and an agent-skills commit on 1 October 2026 (30). Five dated changelog entries between 8 July and 20 September 2026 (20). Public changelog, a Discord community, email and in-app chat, with no public issue tracker for the API (10). Listed in the official MCP registry as `com.typefully/mcp`, active since 1 September 2026 (15). The skill repository runs a test workflow, and there is no SDK to judge (7).

### Security & auth: Repostit +6

- Repostit (44): OAuth with dynamic client registration and no client secret, per Repostit's Claude connector page, with offline access that renews for up to 90 days from the first authorisation; social-network sign-ins stay in Repostit and provider tokens are not passed to the assistant. PKCE S256 is stated only in third-party metadata. No scopes are named; the page lists the permissions shown at sign-in (read accounts, create workflows and Briefs, schedule or publish, read delivery status). Revocation is by disconnecting in the client or asking support (18 of 30). A Brief is saved as a draft, Manual workflows wait for approval, and the help article says the assistant shows each step for confirmation before publishing (on ChatGPT 'may ask', depending on the client); no read-only mode, and an activated future-content workflow publishes on its own (8 of 20). Tools return the user's own accounts, Briefs and delivery results and read no comments or messages (10). Assistant-created posts appear under Post with a Comments & Activity log of each step, and Post History shows each network's response; no per-call log for the connector (8 of 15). security.txt did not load through our reader, and no disclosure policy, bug bounty or certification appears on any page read (0 of 20). 18+8+10+8+0 = 44.
- Typefully (38): OAuth on the MCP server with PKCE, dynamic client registration and revocation, and revocable API keys, but one scope `full_access` and keys that carry all of the creating user's permissions (20). The MCP article says the old URL form with the key in `?TYPEFULLY_API_KEY=` still works (-10). Read, write and publish access per social set is inherited from the user, and planned drafts need a confirming call, but no read-only key exists (10). Responses hold the workspace's own drafts and teammates' comments, with no injection guidance found (6). Draft Activity records who changed a draft and labels MCP or API as the source (10). No security.txt, bug bounty or certification was found, only a short security section in the privacy policy (2).

## Score by category

| Category | Weight | Repostit | Typefully | Edge |
| --- | --- | --- | --- | --- |
| Reliability | 16% (20 this run) | 26 | 27 | Typefully +1 |
| Performance | 10%, pending | pending | pending | not scored in this run |
| Schema & documentation | 13% (16.2 this run) | 27 | 83 | Typefully +56 |
| Agent ergonomics | 13% (16.2 this run) | 25 | 62 | Typefully +37 |
| Security & auth | 14% (17.5 this run) | 44 | 38 | Repostit +6 |
| Payments & pricing | 10% (12.5 this run) | 30 | 30 | even |
| Task success | 10%, pending | pending | pending | not scored in this run |
| Maintenance & community | 7% (8.8 this run) | 56 | 82 | Typefully +26 |
| Transparency & trust | 7% (8.8 this run) | 29 | 66 | Typefully +37 |
| Negative events | ≤15 | 0 | -3 | |
| **Total** | | **32.5 · F** | **49.3 · D** | |

## Facts side by side

| Fact | Repostit | Typefully |
| --- | --- | --- |
| Kind | Platform | HTTP API |
| Vendor | Repostit LLC | Typefully Inc. |
| Hosted endpoint | `https://repostit-mcp-cfjcykgqda-uc.a.run.app/mcp/v1.1` | `https://api.typefully.com` |
| Transports | Streamable HTTP | HTTP, Streamable HTTP |
| Auth | OAuth | OAuth or key |
| Pricing | Paid | Freemium |
| x402 | no | no |
| Licence | Proprietary hosted service under Repostit's terms of service (effective 7 March 2025), which name Repostit LLC of Newark, Delaware. No source, SDK or package is published | Proprietary service under Typefully's Terms of Service. The Agent Skill and its CLI script in typefully/agent-skills are MIT |
| Read-only variant documented | no | no |
| llms.txt | yes | yes |
| MCP registry | `io.github.hax98/repostit` | `com.typefully/mcp` |
| Last release | 2026-10-23 | 2026-10-01 |
| Terms last updated | not read yet | 2026-02-23 |
| Privacy policy last updated | not read yet | no date given |
| Customer content may train models | not read yet | not found in the text |
| Terms restrict automated access | not read yet | yes |
| Terms restrict benchmarking | not read yet | not found in the text |
| Terms or service can change without notice | not read yet | yes |
| Arbitration or class-action waiver | not read yet | not found in the text |

## How the scores have moved

### Repostit

No score change since it was listed.

### Typefully

- +0.2 on 27 October 2026: checklist

## Verdicts

**Repostit.** An OAuth connector lets an agent list accounts, draft and schedule posts, build reposting workflows and check delivery, with drafts saved before anything publishes and a 14-day trial without a card. Tool names, schemas, scopes and error shapes are unpublished, the endpoint sits on a Cloud Run host, and there is no status page, public API or named subprocessor.

**Typefully.** A public OpenAPI 3.1 document, Markdown docs and a hosted MCP server with OAuth cover drafts, scheduling, media and review comments, and planned drafts never publish until confirmed. Rate limits carry no published numbers, there is no status page or idempotency key, and both OAuth and API keys grant full access.

## Before you call either

### Repostit

1. List connected accounts first and confirm names and platforms with the user before choosing destinations; a missing account must be connected in Repostit Connections, not through the agent.
2. Pass media as a downloadable HTTPS URL, never a local filename, and give every scheduled time with an explicit timezone.
3. Treat Queued as accepted, not published; check delivery status and public links before reporting success, and check the existing result before any retry to avoid duplicate posts.
4. Creating a future-content workflow makes it publish on its own once active; prefer existing-content workflows or Manual mode when the user wants to review first.
5. On the trial, each destination counts as one of 10 posts and Basic allows 10 video posts a month; ask before fanning one Brief out to many networks.

### Typefully

1. Call `GET /v2/social-sets` first. Every draft, media, tag and queue path needs a `social_set_id`
2. Use `plan_at` or omit both dates to leave a draft unpublished. `publish_at: "now"` posts at once and is asynchronous
3. Do not retry a failed publish blindly. Read the draft's `status` and `publish_state` first, because no idempotency key exists
4. Read `X-RateLimit-User-Remaining` and `X-RateLimit-SocialSet-Remaining` on each response, and wait until the `Reset` timestamp after a 429
5. Send the key only as `Authorization: Bearer`. The legacy MCP URL with `?TYPEFULLY_API_KEY=` still works and puts the key in logs

## Tutorials

- [Draft a two-post thread with Typefully](https://www.anchorterminal.com/tutorials/typefully/), for Typefully

## Questions

### Which is better for AI agents, Repostit or Typefully?

Typefully scores 49.3 (D) on agent readiness against Repostit's 32.5 (F), and leads in 5 of 7 scored categories. Repostit leads on security & auth.

### Can an agent call Repostit and Typefully without installing anything?

Yes. Repostit has a hosted endpoint at https://repostit-mcp-cfjcykgqda-uc.a.run.app/mcp/v1.1 and Typefully at https://api.typefully.com.


## For agents

- This comparison as JSON: https://www.anchorterminal.com/compare/repostit-vs-typefully.json, and with the fewest tokens: https://www.anchorterminal.com/compare/repostit-vs-typefully.min.md
- Over MCP at https://www.anchorterminal.com/mcp (no key): `compare_tools {"a": "repostit", "b": "typefully"}`. From a terminal: `anchor compare repostit typefully`
- Each listing in full: https://www.anchorterminal.com/api/v1/tools/repostit.json and https://www.anchorterminal.com/api/v1/tools/typefully.json

## Other comparisons with Repostit or Typefully

- [Ayrshare API + MCP vs Repostit](https://www.anchorterminal.com/compare/ayrshare-vs-repostit.md)
- [Ayrshare API + MCP vs Typefully](https://www.anchorterminal.com/compare/ayrshare-vs-typefully.md)
- [Blotato vs Repostit](https://www.anchorterminal.com/compare/blotato-vs-repostit.md)
- [Blotato vs Typefully](https://www.anchorterminal.com/compare/blotato-vs-typefully.md)
- [Buffer API + MCP vs Repostit](https://www.anchorterminal.com/compare/buffer-vs-repostit.md)
- [Buffer API + MCP vs Typefully](https://www.anchorterminal.com/compare/buffer-vs-typefully.md)
- [bundle.social vs Repostit](https://www.anchorterminal.com/compare/bundle-social-vs-repostit.md)
- [bundle.social vs Typefully](https://www.anchorterminal.com/compare/bundle-social-vs-typefully.md)
- [Zernio (formerly Late) API + MCP vs Repostit](https://www.anchorterminal.com/compare/late-vs-repostit.md)
- [Zernio (formerly Late) API + MCP vs Typefully](https://www.anchorterminal.com/compare/late-vs-typefully.md)
- [Metricool API + MCP vs Repostit](https://www.anchorterminal.com/compare/metricool-vs-repostit.md)
- [Metricool API + MCP vs Typefully](https://www.anchorterminal.com/compare/metricool-vs-typefully.md)
- [Mixpost API + MCP vs Repostit](https://www.anchorterminal.com/compare/mixpost-vs-repostit.md)
- [Mixpost API + MCP vs Typefully](https://www.anchorterminal.com/compare/mixpost-vs-typefully.md)
- [OneUp API + MCP vs Repostit](https://www.anchorterminal.com/compare/oneup-vs-repostit.md)
- [OneUp API + MCP vs Typefully](https://www.anchorterminal.com/compare/oneup-vs-typefully.md)
- [Post Bridge API + MCP vs Repostit](https://www.anchorterminal.com/compare/post-bridge-vs-repostit.md)
- [Post Bridge API + MCP vs Typefully](https://www.anchorterminal.com/compare/post-bridge-vs-typefully.md)
- [Postiz API + MCP vs Repostit](https://www.anchorterminal.com/compare/postiz-vs-repostit.md)
- [Postiz API + MCP vs Typefully](https://www.anchorterminal.com/compare/postiz-vs-typefully.md)
- [Publer API + MCP vs Repostit](https://www.anchorterminal.com/compare/publer-vs-repostit.md)
- [Publer API + MCP vs Typefully](https://www.anchorterminal.com/compare/publer-vs-typefully.md)
- [Repostit vs Upload-Post API + MCP](https://www.anchorterminal.com/compare/repostit-vs-upload-post.md)
- [Typefully vs Upload-Post API + MCP](https://www.anchorterminal.com/compare/typefully-vs-upload-post.md)
- [Repostit vs SocialCannon](https://www.anchorterminal.com/compare/repostit-vs-socialcannon.md)
- [SocialCannon vs Typefully](https://www.anchorterminal.com/compare/socialcannon-vs-typefully.md)

---

Query this catalogue directly over MCP: https://www.anchorterminal.com/mcp (Streamable HTTP, no key). How: https://www.anchorterminal.com/agents/#mcp