Skip to content

Add the team's personal-agent picks, launch Claude Duet, and link Every articles - #43

Merged
kieranklaassen merged 9 commits into
mainfrom
feat/import-personal-agent-picks
Oct 21, 2026
Merged

kieranklaassen merged 9 commits into
mainfrom
feat/import-personal-agent-picks

Conversation

@kieranklaassen

@kieranklaassen kieranklaassen commented Oct 21, 2026 •

Copy link
Copy Markdown
Collaborator

Adds the Every team's personal-agent survey to Toolbox.

  • Catalog: Codex is also a personal agent; Loop is in as Unknown (unannounced-agent) so it can be renamed in config/catalog.yml when it launches.
  • bin/rails toolbox:import_personal_agents: creates the 21 surveyed members (onboarded, private, so counted anonymously and never named) and sets their ranked picks. A first Every sign-in adopts each row by email. Idempotent: anyone with personal-agent picks is skipped.
  • Not imported: Natalia's own orchestrator (not in the catalog). Mike Taylor is added with no picks. Tyler's tie is stored as Grok Bot 1, Unknown 2.

🤖 Generated with Claude Code


Note

Medium Risk
Production rake tasks create users and mutate picks; launch_duet refuses until the catalog rename, but mistaken deploy order or re-runs need ops care. Article/link allowlisting is security-sensitive but reuses the existing Vibe Check host rules.

Overview
Imports the Every team’s October 2026 personal-agent survey via toolbox:import_personal_agents: creates missing @every.to members (onboarded, link-visible), sets ranked picks idempotently, and documents toolbox:launch_duet to backfill Claude Opus 5.5 on Duet picks after the catalog names unannounced-agent as Claude Duet. Catalog adds Codex to personal agents and the Duet tool entry (slug unchanged for existing picks).

Kind pages rename “Takes on these models” to “Read on Every” and broaden takes to catalog articles plus Vibe Checks, with a shared EveryLink allowlist and Toolbox UTM tagging. The UI and get_team_rankings expose { tag, title, item, url } instead of model-only Vibe Check rows. articles in config/catalog.yml is wired through Article (currently empty).

Reviewed by Cursor Bugbot for commit e797e40. Bugbot is set up for automated code reviews on this repo. Configure here.

Codex now counts as a personal agent, and Loop joins the catalog as
"Unknown" (slug unannounced-agent) until it launches and can be renamed.
toolbox:import_personal_agents creates the 21 surveyed members, private
and onboarded, and sets their ranked picks without touching anyone who
already chose personal agents.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@cursor

cursor Bot commented Oct 21, 2026

Copy link
Copy Markdown

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: 273ace8f-b8fc-44ff-a375-a628e7dfc82c)

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@cursor

cursor Bot commented Oct 21, 2026

Copy link
Copy Markdown

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: 04ec17a4-172c-4dd7-9c66-6da6829d8784)

kieranklaassen and others added 2 commits October 21, 2026 11:34
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The agent listed as Unknown becomes Loop by Anthropic, running Claude
Opus 5.5 first. Its slug stays, so the picks already made carry over.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@cursor

cursor Bot commented Oct 21, 2026

Copy link
Copy Markdown

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: 6db9e3fc-48f5-4dea-9c18-4ad19d36eb50)

kieranklaassen and others added 5 commits October 21, 2026 11:35
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Anthropic launched it as Claude Duet, built around Claude Opus 5.5.
The launch task is now toolbox:launch_duet.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A kind's takes become one "Read on Every" list: articles set in
config/catalog.yml for that kind, then the Vibe Checks of launched models
ranked there, each tagged article or vibe_check and carrying Toolbox's
UTM so Every sees where readers came from. The Kind page and
get_team_rankings share the list, and the allowlist moves to EveryLink
so both kinds of link are held to the same Every hosts.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Launch Claude Duet, and link Every articles from kinds
@cursor

cursor Bot commented Oct 21, 2026

Copy link
Copy Markdown

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: e9d407db-f90c-48a1-ba9b-1ab26128d1fd)

@kieranklaassen kieranklaassen changed the title Add the team's personal-agent survey picks Add the team's personal-agent picks, launch Claude Duet, and link Every articles Oct 21, 2026
@kieranklaassen
kieranklaassen merged commit 29415f5 into main Oct 21, 2026
5 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant