// SAMPLES

What subscribers get

Real entries, pulled from issues we already published — not mockups. Every one links to the issue it ran in, so you can read what surrounded it and judge the coverage rather than the excerpt. Compare plans →

// TAIL — the daily release firehose

A real entry from issue 001

Published August 19, 2026 · read that issue

TAIL 001 paid · every weekday

One point release, sixteen changes — two new environment variables named exactly as you would export them, a keybinding action, a GitLab MR badge in the statusline, and the setting that turns auto-resume off. A daily agent release nobody writes up, read to the depth you would need to actually change how you run it.

Everything that shipped across the watchlist that day, and the features each release brings, with runnable commands where they help. A quiet day publishes nothing rather than padding an issue.

Claude Code ai-coding-agents

16 capabilities 2 worked examples

Claude Code v2.1.234 adds CLAUDE_CODE_PROJECT_DIR_NAME, selection:clear keybinding, GitLab MR badge, auto-resume at usage limit, and NT-namespace path rejection.

Adds the optional CLAUDE_CODE_PROJECT_DIR_NAME environment variable, letting hosts that give each session its own config directory choose a short name for the per-project transcript directory.
Adds the CLAUDE_CODE_GOAL_CHECKIN_MINUTES environment variable (set to 0 to opt out) — when background tasks keep a /goal waiting 30+ minutes, Claude now checks in instead of waiting indefinitely.
Adds the selection:clear keybinding action, which can be bound to a key to clear an in-app text selection; also works in the agents view.
Adds a GitLab merge request badge to the footer and statusline: repos with a GitLab remote and an authenticated glab CLI show MR !N with draft/pending/green states.
Session auto-resume: Claude Code now continues a session automatically when a claude.ai usage limit resets; disable in /config under 'Continue automatically at usage limit'.
/permissions can now be opened while Claude is actively working — rule changes apply immediately to the rest of the current turn.
/add-dir <path> can now be invoked mid-turn; /add-dir, /autocompact, /theme, /help, /config, and /advisor dialogs all open mid-turn in the fullscreen TUI.
/goal clears itself with a notice when a turn dies on an unrecoverable error (revoked auth, exhausted credit balance, or context overflow) instead of staying armed.

…and 8 more, in this one entry.

Worked example — carried in the issue

Give a multi-user host a predictable, short transcript directory name per project instead of a generated one.

CLAUDE_CODE_PROJECT_DIR_NAME=my-api-service claude

0 more entries in that issue. Every Tail issue is free to read on the site.

Read issue 001 in full → ▷ See plans

// HEAD — the week's top picks

A real entry from issue 001

Published August 21, 2026 · read that issue

HEAD 001 free · fridays

Three APIs left beta in one week — Files, Skills and Enterprise user management — which means three beta headers you can now delete, new pagination and filter parameters, and a 1 TB storage ceiling. Buried in a release-notes page, and the kind of change that quietly breaks a client still sending the old header.

The week's top picks — the releases we'd flag first — each with a short "why it matters". If nothing clears the bar, no issue ships.

Anthropic ai-infrastructure

23 capabilities 5 breaking changes 2 worked examples

Claude API reaches GA for Files, Skills, and Enterprise user management; Managed Agents gains budgets, advisors, memory stores, domain restrictions, and more.

Admin API user-management endpoints (members, invites, groups, custom roles) for Claude Enterprise organizations are now GA; the anthropic-beta: ce-user-management-2026-07-13 beta header is no longer required on group and custom-role requests.
Files API is now GA on /v1/files; the files-api-2025-04-14 beta header is no longer required. GA response format adds expires_in_seconds (upload) and expires_at (file objects), page and next_page pagination, and an ids[] filter on list requests. Storage limit is 1 TB per organization; rate limit is 500 requests per minute.
Agent Skills and the Skills API (/v1/skills) are now GA; the skills-2025-10-02 beta header is no longer required, including for Messages API requests that load Skills through the container parameter.
Adds allowed_domains and blocked_domains controls on web_search and web_fetch tool entries in the agent_toolset_20260401 configs array for Claude Managed Agents; web_fetch also accepts max_content_tokens and web_search accepts user_location.
Claude Managed Agents sessions running in a self-hosted sandbox can now attach memory stores; Python, TypeScript, and Go SDK workers download each store to its mount_path and sync changes back.
Redesigned session viewer in the Claude Console adds a timeline minimap, transcript grouped by model request, and an Inspector panel covering session details and cost, raw events, per-tool statistics, mounted resources, and per-thread activity.
Workbench is now Playground at platform.claude.com/playground; supports every Messages API parameter, includes feature templates (code execution, web search), and shows the full SDK request and API response for each run.
Compliance API (beta for Claude Enterprise) adds GET /v1/compliance/apps/sessions/local to list local Cowork and Claude Code sessions, GET /v1/compliance/apps/sessions/local/{session_id} for session metadata, and GET /v1/compliance/apps/sessions/local/{session_id}/messages for transcripts, using the existing Compliance Access Key with read:compliance_user_data scope.

…and 15 more, in this one entry.

Worked example — carried in the issue

Upload a file with a TTL and then reference it in a Messages API request without the old beta header.

curl https://api.anthropic.com/v1/files \
  -H 'x-api-key: $ANTHROPIC_API_KEY' \
  -H 'anthropic-version: 2023-06-01' \
  -F '[email protected];type=application/pdf' \
  -F 'expires_in_seconds=86400'

0 more entries in that issue. Every Head issue is free to read on the site.

Read issue 001 in full → ▷ See plans

// CHARTS — ANALYTICS PUBLISHED NOWHERE ELSE

What the market looks like from inside the corpus

Invented tools and invented numbers — an illustration of the view, not a measurement. The real versions are on the analytics page with their labels masked.

illustration

Who is slipping, who is breaking, and where a category is heading — drawn from every release we have collected. The live charts are here — the same rankings, with the names masked.

Who documents what they ship

Release-notes quality, scored 0-100 on the median note.

Without one

▒▒▒▒▒▒▒▒▒▒▒▒ 88 / 100

▒▒▒▒▒▒▒▒▒ 71 / 100

▒▒▒▒▒▒▒▒▒ 46 / 100

▒▒▒▒▒▒▒▒▒ 22 / 100

With a plan

Northwind Scanner 88 / 100

Kestrel Vault 71 / 100

Pinegrove EDR 46 / 100

Halberd Cloud 22 / 100

Whose platform goes down

Hours degraded or down in the last 90 days, from the vendor’s own status page.

Without one

▒▒▒▒▒▒▒▒▒ 41h 20m

▒▒▒▒▒▒▒▒ 24h 05m

▒▒▒▒▒▒▒▒▒ 8h 30m

▒▒▒▒▒▒▒▒▒▒▒▒ 2h 45m

With a plan

Halberd Cloud 41h 20m

Vellum SIEM 24h 05m

Kestrel Vault 8h 30m

Northwind Scanner 2h 45m

Whose fixes don’t hold

Share of fixes that reopen the same failure within a quarter.

Without one

▒▒▒▒▒▒▒▒▒ 38% reopen

▒▒▒▒▒▒▒▒ 25% reopen

▒▒▒▒▒▒▒▒▒ 16% reopen

▒▒▒▒▒▒▒▒▒▒▒▒ 5% reopen

With a plan

Pinegrove EDR 38% reopen

Vellum SIEM 25% reopen

Halberd Cloud 16% reopen

Northwind Scanner 5% reopen

Researcher and Business carry the full analysis, across release-notes quality, platform reliability, category momentum and the defect analysis. ▷ See plans