`, alongside the existing stylesheet and stale-banner regions.
- Workshop deliverables render as `
`.
- Each deliverable carries a kind of `pr`, `artifact` or `other`, with links, parsed out and surfaced next to decisions.
**Evidence**
`stamp-control`
- Area: Artifacts
- Tier: You'll notice
- Useful: 2/5
- Signal: 0/5
### Web sessions can adopt refreshed credentials from the heartbeat
Web sessions can swap in a refreshed credential mid-run instead of waiting for the old one to expire.
**What**
The Claude-Code-on-the-web worker client can pick up a new credential mid-session when the server sends one on a heartbeat, rather than waiting for the existing credential to lapse.
**Details**
- New `adoptRefreshedAuth`: if a heartbeat response carries `refreshed_auth` with `expires_in_seconds`, the worker adopts the credential in place.
- Diagnostics `cli_heartbeat_refreshed_auth_signal`, `cli_heartbeat_refreshed_auth_late_adopted` and `cli_heartbeat_refreshed_auth_adopt_threw`.
- Repeated failure reasons are deduped via `lastAuthRefreshBadReason`.
- Only wired in when `CLAUDE_CODE_ENVIRONMENT_KIND` is unset; otherwise the handler is not passed at all.
**Evidence**
`cli_heartbeat_refreshed_auth_signal`
- Area: Auth
- Names: `CLAUDE_CODE_ENVIRONMENT_KIND`
- Tier: Under the hood
- Useful: 2/5
- Signal: 0/5
### Published PR-review pages carry an approve binding
Published PR review pages now carry an approve action, and republishing can't quietly change it.
**What**
A composed PR review artifact now ships an approve script and a stamp island recording the approve binding, and republishing is checked against it.
**Details**
- The artifact includes an `approveScript` plus a stamp island holding the approve binding.
- Validation rejects a republish that changes the binding.
- All approve-binding validation messages are new in this build.
**Evidence**
`a republish must reuse the published stamp island exactly`
- Area: Artifacts
- Tier: You'll notice
- Useful: 2/5
- Signal: 0/5
## Improvements
### resources/list and prompts/list stop recording an error when the server lacks the method
MCP servers that don't support resource or prompt listing no longer show a bogus error.
**What**
MCP servers that simply do not implement resource or prompt listing no longer surface a discovery error.
**Details**
- Both discovery functions now skip attaching the stored error when the failure is a MethodNotFound JSON-RPC error, matching the behaviour resource templates already had.
**Evidence**
`Ou.MethodNotFound`
- Area: MCP
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### HTTP status codes read from the error's status instead of its code
Fixes MCP re-authentication misfiring by reading the real HTTP status instead of guessing from messages.
**What**
MCP auth-challenge classification is now driven by the HTTP error's status property rather than its code field.
**Details**
- Auth-challenge classification for sse/http/claudeai-proxy connect failures and the ENDPOINT_NOT_FOUND special case prefer `.status`, falling back to `.code`.
- The claudeai-proxy 401/403 re-auth path does the same.
- The GET-stream 404 check no longer matches on the message text.
**Evidence**
`get_stream_404_not_reinit`
- Area: MCP
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Artifact live-update sockets send keepalive pings
Live artifact updates now ping every 25 seconds so long sessions stop getting dropped.
**What**
Long-lived artifact subscriptions should stop being dropped by idle timeouts.
**Details**
- the live-subscribe websocket is opened with a keepalive config sending a "ping" frame every 25s (iV_ = 25000)
- "ping" and "hb" are treated as open frames
- the kill path now also aborts the subscription controller when the user stops the task
**Evidence**
`openFrames: ["ping", "hb"]`
- Area: Artifacts
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Background-session prompt rewritten around committing and a final report
Background jobs now commit and push when possible and end with an actionable report instead of always opening a draft PR.
**What**
The background job system prompt drops the "shipping is part of the task" framing and the mandatory draft PR.
**Details**
- says to commit before finishing and to push if there is a remote
- defers to user or CLAUDE.md instructions that reserve git
- the draft PR is now conditional rather than required
- adds a closing instruction to end with an actionable report, with a carve-out for subagents
**Evidence**
`End the job with a report the user can act on: what you did, where it lives`
- Area: Sessions
- Names: `CLAUDE.md`
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Forked background sessions get clearer worktree guidance
Forked background sessions are told to stay in their own worktree and never touch the original session's branch.
**What**
The system prompt appended to a forked session that keeps the parent alive was rewritten to be explicit about isolation.
**Details**
- names the tool to use for isolation
- tells the fork never to enter the original session's worktree and explains why not to check out the original's branch
- adds a new "own-worktree" case for forking out of the user's own checkout
- the line is skipped if the same text is already present in appendSystemPrompt
- only added for forks that keep the parent session alive
**Evidence**
`so your edits don't land where the original session is editing.`
- Area: Sessions
- Names: `appendSystemPrompt`
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### MCP slash-command prompts extracted, now abortable
You can now cancel a running MCP prompt slash command without seeing a bogus error message.
**What**
Running an MCP prompt as a slash command can now be aborted cleanly, and aborting no longer looks like a failure.
**Details**
- The prompt-command construction moved into its own function.
- Fetching the prompt passes the command's abort signal and a context label to the client connection.
- An aborted prompt no longer logs mcp_get_prompt_failed and no longer surfaces an "Error running command" notice; it rethrows.
**Evidence**
`context: "MCP prompt command"`
- Area: MCP
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Cloud review: better messages for detached HEAD and wrong base branch
Cloud review now tells you to create a branch on detached HEAD instead of failing with a vague error.
**What**
Cloud review failures on a checkout with no branches now return a dedicated `no_merge_base` error telling you to create a branch first, and the "shares no history with HEAD" message suggests passing the real base branch explicitly.
**Details**
- detached-HEAD checkouts previously fell through to a generic bundling failure
- the no-shared-history path now names the likely cause rather than only reporting it
**Evidence**
`Your checkout has no branches (detached HEAD only), which cloud review can't bundle.`
- Area: Code Review
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Plugin install now tells you when higher-precedence settings keep it disabled
Plugin install now tells you when a higher-priority settings file is what's keeping the plugin disabled.
**What**
After a plugin install, Claude Code checks the settings layers above the install scope for `enabledPlugins[id]` set to false and reports that case with its own message.
**Details**
- walks each higher-precedence settings layer rather than only reading the install scope
- distinguished in the install result from the ordinary "disabled by default" outcome, with a distinct message
**Evidence**
`This plugin is disabled in your settings`
- Area: Plugins
- Names: `enabledPlugins`
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### MCP error classification recognises typed HTTP and OAuth errors
MCP auth and HTTP failures are classified more accurately, and cancelled tool calls no longer trigger slow-call warnings.
**What**
Tool-call failure handling now reads status and OAuth codes off typed error classes instead of guessing from a loose `code` property, and stops warning about slow calls that were simply aborted.
**Details**
- status is derived from typed HTTP error classes; the old path only used `code`, and only when it was a number
- OAuth error codes come from a typed error object
- the "Tool 'X' failed after Ns" warning is suppressed when the call was aborted
- a full list of MCP and OAuth codes is enumerated for classification, including invalid_grant, AUTH_HEADER_REJECTED, CLIENT_HTTP_FORBIDDEN and ERA_NEGOTIATION_FAILED
**Evidence**
`CLIENT_HTTP_FAILED_TO_TERMINATE_SESSION`
- Area: MCP
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Browser tabs: Claude is told to close tabs it opened
Claude now closes browser tabs it opened, so you get fewer stray tabs left behind.
**What**
The `tabs_create_mcp` description now includes explicit cleanup instructions, so browser sessions should accumulate fewer stray tabs.
**Details**
- each created tab should be closed with `tabs_close_mcp` once the work is done
- a tab is left open only when the user asked for it
**Evidence**
`Tabs you create are yours to clean up`
- Area: Chrome & Browser
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Reconnecting a disabled MCP server tells you to enable it
Reconnecting a disabled MCP server now just tells you to enable it first.
**What**
The /mcp reconnect flow checks the server's status first, and a disabled server reports that it needs enabling instead of running the reconnect state machine.
**Details**
- status check happens before any reconnect attempt, so no connection cycle is started for a disabled server
- the message points you back at /mcp to enable the server
- other statuses still go through the normal reconnect path
**Evidence**
`enable it in /mcp first`
- Area: MCP
- Names: `/mcp`
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### read_page_data needs a human approval in plan mode
In plan mode, reading artifact page data now needs your actual approval, not an auto-approval.
**What**
Approval for the artifact read_page_data action is tracked in two separate app-state flags, and in plan mode the tool auto-allows only when the human-approved flag is set.
**Details**
- a plan-mode auto-approval no longer carries over into the human-approved flag
- checkLiveEditPermissions now receives the plan-consent denier and the permission context
- outside plan mode the existing approval flag still applies
**Evidence**
`artifactReadPageDataHumanApproved`
- Area: Permissions
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Windows sandbox helper: hidden windows and real timeout errors
On Windows, sandbox helper windows stay hidden and timeouts now tell you a UAC prompt may be open.
**What**
srt-win invocations spawn with windowsHide set, and a spawn that returns ETIMEDOUT now raises a typed timeout error instead of a generic spawn failure.
**Details**
- the timeout error carries the full argv and the elapsed milliseconds
- the install path maps that code onto a message about the UAC prompt possibly still being open
- Windows only, and only on the srt-win sandbox path
**Evidence**
`srt_win_timeout`
- Area: Windows
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Spend-limit message distinguishes an org spend cap
Hitting a spend cap now says whether it's your individual limit rather than always the org's.
**What**
When the API reports overageDisabledReason "org_spend_cap_reached", the limit banner names your individual spend limit instead of always naming the org's monthly limit.
**Details**
- the banner reads "You've hit your individual spend limit" for that reason
- the rejected-overage branch uses "individual usage limit"
- both phrases are new in this build; other reasons keep the org monthly wording
**Evidence**
`individual spend limit`
- Area: Usage & Limits
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Memory recall index is prewarmed at startup
Claude warms the memory recall index at startup so your first recall is faster.
**What**
A startup task warms the recall and memory index in the background so the first recall does not pay the indexing cost.
**Details**
- Invoked once from both boot paths.
- Bails out if the invocation looks like a resume (`--resume`, `--continue`, `--from-pr`, or a short flag containing c or r), or if index exclusions are already loaded.
- Only proceeds when auto-memory, memory recall and index recall are all enabled, and waits for the org-memory decision to settle when multi-store sync is in play.
- Failures are swallowed with a debug line rather than surfaced.
**Evidence**
`maybePrewarmRecallIndex`, `recall prewarm skipped: `
- Area: Memory
- Names: `--resume`, `--continue`, `--from-pr`
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### MCP enable/disable failures report the underlying error
Turning an MCP server on or off now shows the real error instead of a vague message.
**What**
Failing to toggle an MCP server now shows the actual error instead of a generic "it may have been removed" message.
**Details**
- Applies to both the /mcp slash-command path and the plugin/MCP toggle UI.
- Toggling a server from the manager surfaces the error inline rather than discarding it.
- The failure text is assembled at runtime from the action, the server name and the error.
**Evidence**
`Failed to `
- Area: MCP
- Names: `/mcp`
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### MCP OAuth refresh retries on error codes and preserves issuer and redirect URI
MCP OAuth token refresh retries smarter and keeps your saved redirect URI across reconnects.
**What**
Token-refresh retry decisions are made from OAuth error codes rather than exception classes, and stored client registrations keep more of the discovered configuration.
**Details**
- Retries on server_error, temporarily_unavailable, too_many_requests, or any code not in the known non-retryable set.
- The analytics branch reads `code` instead of `errorCode`.
- Registrations keep the discovered `issuer` and reuse the previously stored redirectUri when the client_id matches what is on disk.
- Discovery state is cached on the flow object.
**Evidence**
`mcp_oauth_refresh_unauthorized_client`
- Area: MCP
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Installed-but-inactive plugins say why
Inactive plugins now tell you whether they ship disabled or your settings disabled them.
**What**
The message for a plugin that is installed but not running now tells you whether it ships disabled or whether your own settings turned it off.
**Details**
- Applies to both the install result message and the plugin-not-enabled diagnostic.
- Distinguishes plugins disabled by default from plugins disabled in your settings.
- The enable hint still follows either variant.
**Evidence**
`. This plugin is disabled in your settings`
- Area: Plugins
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Auto mode always sees same-turn sibling tool uses
Permission checks in auto mode now always see other tool calls from the same turn.
**What**
Sibling tool uses from the same turn are now always folded into the messages passed to permission evaluation.
**Details**
- Previously `sameTurnToolUses` was included only when a gate function backed by `CLAUDE_CODE_AUTO_MODE_SIBLING_CONTEXT` returned true.
- That gate and its environment variable are both removed from the build.
**Evidence**
`sameTurnToolUses`
- Area: Permissions
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Auto mode exit message is tailored, and subagents no longer get it
The auto mode exit notice now matches how you configured it, and subagents no longer see it.
**What**
The notice shown when auto mode ends now varies with how auto mode was configured, and it is suppressed for subagents.
**Details**
- The attachment lookup returns the attachment rather than a boolean, so the exit attachment carries `bashFirst` and `steerOnly`.
- The bash-first variant appends " Resume using the dedicated tools for file reads, searches, and edits."
- Subagents get nothing: `if (t.agentId) return []`.
**Evidence**
`You have exited auto mode.`
- Area: Auto Mode
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### MCP elicitation schemas are validated with a clear error
A bad MCP elicitation schema now shows a clear error instead of a misleading reconnect message.
**What**
A malformed `requestedSchema` in an MCP elicitation now produces an InvalidParams error naming the constraint, rather than a misleading reconnect message.
**Details**
- parse failure on `requestedSchema` is converted into an MCP InvalidParams error explaining the constraint
- the completion notification handler is registered against the literal method name `notifications/elicitation/complete`
**Evidence**
`notifications/elicitation/complete`
- Area: MCP
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### MCP OAuth tracks the issuer through the callback and token exchange
MCP OAuth now tracks which authorization server issued your token through login, refresh, and storage.
**What**
The OAuth flow now carries the authorization server issuer from the callback into the token exchange and into stored credentials.
**Details**
- both the local callback listener and the manual callback-URL path read the `iss` query parameter and pass it into the token exchange
- the local callback server resolves `{ code, iss }` instead of just a code
- stored and served client information and the token response carry an `issuer` field, and the token provider returns `{ ...a, issuer: n?.issuer }` on both the refresh and the silent-exchange paths
- `_flowDiscoveryState` caches the discovered authorization-server metadata on the provider instance
- revoke calls a cleanup helper before reporting `mcp_oauth_revoke`
- new device-code error sets were added, including `authorization_pending`
**Evidence**
`authorization_pending`, `_flowDiscoveryState`
- Area: MCP
- Tier: Under the hood
- Useful: 3/5
- Signal: 0/5
### Org spend-cap limits get their own wording
Hitting an org spend cap now shows its own message with links matched to your account type.
**What**
Hitting an organization spend cap now reads differently from hitting a monthly usage limit, with links matched to the account type.
**Details**
- `org_spend_cap_reached` is handled separately from the generic "org's monthly usage limit" message
- team and enterprise accounts get a spend-limit message pointing at /usage-credits or claude.ai/admin-settings/usage
**Evidence**
`org_spend_cap_reached`
- Area: Usage & Limits
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### MCP OAuth handles the RFC 9207 issuer parameter
MCP OAuth logins now verify the issuer and give clearer errors when client registration is rejected.
**What**
Loopback and manual-URL OAuth callbacks now read `iss` from the query string and carry it into token exchange, and saved MCP tokens record the issuer.
**Details**
- Token exchange receives `authorizationCode: d.code, iss: d.iss`.
- Saved tokens gain an `issuer` field; an unexpected-issuer check runs when the authorization server advertises iss support.
- New failure classification `"dcr_rejected"` separates a rejected dynamic client registration from the existing `"dcr_failed"`.
- OAuth error matching moved off message and `errorCode` sniffing onto typed SDK errors read via `.code`, covering `invalid_client`, `unauthorized_client` and InvalidGrant.
**Evidence**
`dcr_rejected`
- Area: MCP
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Rewritten inline /mcp replies for cross-session enable/disable
/mcp now tells you when another session already enabled or disabled a server, and how to make your change stick.
**What**
The inline /mcp command now says explicitly when another session already re-enabled or disabled a server, and tells you how to make your change stick.
**Details**
- New messages for the cross-session cases, including a two-step recovery instruction.
- New precondition: three controls must be available rather than two.
- Per-server error text now comes from a shared formatter instead of the old inline "may have been removed" wording.
**Evidence**
`makes it stick. Left alone, it connects on the next launch.`
- Area: MCP
- Names: `/mcp`
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### /mcp explains cross-session enable/disable drift
/mcp gains clearer messages for cross-session enable/disable conflicts and servers blocked by managed policy.
**What**
New messages cover the cases where an enable or disable in one session conflicts with what another session has already done.
**Details**
- Distinct text for a server another session disabled, and for one another session re-enabled so your disable did not persist.
- Separate text for a server blocked by managed policy.
- None of these strings exist in the previous build.
**Evidence**
`was disabled in another session`
- Area: MCP
- Names: `/mcp`
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### /plugin reload checks prompt-cache impact
Activating plugins from the /plugin dialog now warns you before it invalidates your conversation's prompt cache.
**What**
Activating plugins from the plugin management dialog now goes through the same prompt-cache check that /reload-plugins uses, so an install does not silently invalidate the conversation's prompt cache.
**Details**
- The dialog is passed `getReloadCacheImpactOptions`, carrying the model plus dynamic MCP config.
- An install-triggered activation can now bail out with a `cache_impact` outcome.
- The equivalent /reload-plugins warning text already existed.
**Evidence**
`getReloadCacheImpactOptions`
- Area: Plugins
- Names: `/plugin`, `/reload-plugins`
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Plugin validation warns about Claude Desktop marketplace compatibility
Plugin validation warns when marketplace or plugin names would be rejected by Claude Desktop's managed sync.
**What**
Marketplace and plugin names that Claude Code accepts but Claude Desktop's managed sync would reject now produce a warning at validation time.
**Details**
- Names are checked against Claude Desktop's stricter pattern `/^[A-Za-z0-9][A-Za-z0-9._-]{0,127}$/`: alphanumeric start, then letters, digits, `.`, `_` or `-`, up to 128 characters.
- Reserved marketplace names are `org`, `org-provisioned` and `unknown`.
- The warning explains that the managed `allowedPluginMarketplaces` sync will reject the whole marketplace, or drop the individual plugin entry.
- Claude Code itself still accepts these names.
**Evidence**
`is reserved in Claude Desktop`, `is reserved in Claude Desktop. Claude Code accepts it, but the Claude Desktop managed marketplace sync (allowedPluginMarketplaces) will reject the whole marketplace.`
- Area: Plugins
- Names: `allowedPluginMarketplaces`
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Artifact credentials and artifact base URLs are stripped from child process environments
Artifact tokens and artifact base URLs no longer leak into processes Claude spawns.
**What**
The environment handed to spawned processes now drops the artifacts API token and any artifact base URL variable, alongside the OAuth token and subscription vars it already removed.
**Details**
- A new predicate matches any variable that starts with `CLAUDE_CODE_ARTIFACT` and ends with `_BASE_URL`, and those are deleted.
- The presence of such a variable alone is now enough to force the sanitizing path instead of returning `process.env` untouched.
- The artifacts token was added to the sensitive-variable list alongside ANTHROPIC_API_KEY and CLAUDE_CODE_OAUTH_TOKEN, and to the set stripped from config-derived env.
- Applies to everyone on this build.
**Evidence**
`return e.startsWith("CLAUDE_CODE_ARTIFACT") && e.endsWith("_BASE_URL");`
- Area: Permissions
- Names: `CLAUDE_CODE_OAUTH_TOKEN`, `ANTHROPIC_API_KEY`
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### MCP task restore dials lazily-cached servers instead of giving up
Resuming a long-running MCP task after a restart now reconnects cached servers instead of failing.
**What**
Resuming a long-running MCP task after a restart now connects a server that is in the "cached" state on demand, rather than treating it as unavailable.
**Details**
- the dial happens within the remaining restore budget, with distinct failure reasons when no session factory is wired and when the dial itself fails
- restore also detects a server that reconnected on a modern-era protocol revision, which carries no tasks support, and reports the task as unresumable instead of hanging in the poll loop
- the "cached" state comes from the new MCP discovery cache, gated on `tengu_mcp_discovery_cache` with an in-source fallback of true
**Evidence**
`server '${e.serverName}' has no MCP session wired`
- Flag `tengu_mcp_discovery_cache`: Not enough to say (read for one account on one subscription tier against v2.1.221; this account: no value returned, anonymous baseline: no value returned, compiled default: off) These values were read against a different version of Claude Code, so treat them as the nearest reading available instead of one taken on this release.
- Area: MCP
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Subagent frontmatter MCP servers connect through the shared bounded dial
A hung MCP server listed in an agent's frontmatter can no longer stall subagent startup.
**What**
MCP servers declared in an agent's frontmatter are connected through the shared dial helper with a timeout and abort signal, so a hung server can no longer stall subagent startup indefinitely.
**Details**
- previously the code awaited `connectToServer` directly with no bound
- on timeout it logs a warning and returns no client
- if the server entry was newly created, the connection is still cleaned up should it land after the timeout
**Evidence**
`subagent MCP server connect`
- Area: MCP
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### ListMcpResources works against cached servers and connects on demand
Listing MCP resources now includes cached servers, connecting them on demand instead of skipping them.
**What**
The resource-listing tool now returns resources from servers that are only in the discovery cache, connecting them as part of the call.
**Details**
- the old `i.type !== "connected"` bail is gone; any connected-or-cached client declaring `capabilities?.resources` is accepted
- it then calls `ensureConnectedClient` with the tool's abort signal and a context label before fetching
- failures are logged per server, except aborts, which are silent and simply contribute no resources
**Evidence**
`MCP resource listing`
- Area: MCP
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Batch plugin install distinguishes disabled-by-default from disabled-in-settings
Batch plugin installs now tell you which plugins are off by default versus off in your settings.
**What**
The multi-plugin install summary tells you why each installed plugin is not running, and the closing line counts only the plugins that actually came up.
**Details**
- Two separate lists: " Disabled by default — enable in /plugin: " for manifest defaults and " Disabled in your settings: " for the user's own settings.
- The active count decides between "Plugins are now active.", "Run /reload-plugins to activate." and " Some plugins couldn't be loaded — see /plugin for details.".
- The single-plugin screen carries matching `installedDisabled` and `installedDisabledByDefault` view-state fields, derived from the `enabledPlugins` setting and `manifest.defaultEnabled`.
**Evidence**
` Disabled in your settings: `
- Area: Plugins
- Names: `/plugin`, `/reload-plugins`
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Workshop skill: two-stage publish and cheaper template reads
Workshop pages now publish a first shareable version faster, with decisions added in a second pass.
**What**
The workshop skill prompt was rewritten so the first shareable page goes up sooner and the template is read with far fewer tokens.
**Details**
- One setup turn: load artifact-capabilities, load artifact-diagramming, and `cp` the template.
- Then two ranged Reads covering only the authored parts, with the prompt stating "The template is 2,919 lines" and naming which line ranges are fixed template bytes.
- Mandates "Publish twice — the first page goes up fast.": an opening version with no decisions and an empty island, then a second publish adding the decision cards.
- The artifact-design skill is explicitly not loaded on this lane.
**Evidence**
`**Set up in one turn, read the template in one more.**`
- Area: Artifacts
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Workshop page layout leads with the working draft
Workshop pages now lead with the working draft above the decisions, with richer diagrams each round.
**What**
Both bundled workshop templates now put the working draft at the top of the page as the hero, ahead of the decisions.
**Details**
- "Working draft" moved above "What we're deciding" in both templates.
- The direct-HTML template adds a `.draft` rule: accent left border, surface background, full-width figure SVG.
- The direct-HTML template also ships a much richer sample main diagram and per-decision diagram.
- The skill text now requires the main diagram be redrawn every round, and per-decision diagrams to show the actual mechanism rather than one labeled box per option.
**Evidence**
`The working draft is the page hero`
- Area: Artifacts
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Artifact comment and reply tool previews show the target and its sharing state
Artifact comment and reply prompts now show which artifact you're acting on and whether it's shared.
**What**
The permission and preview renderer now describes what a `comments` or `reply` call will act on, rather than showing a bare action.
**Details**
- Prints the resolved artifact address, falling back to `(unrecognized address)` when it cannot be resolved.
- In verbose mode, appends the share visibility, or `(share status unconfirmed)` when the share probe failed.
- Verbose mode also shows the reply text for `reply`.
**Evidence**
` (share status unconfirmed)`
- Area: Artifacts
- Tier: You'll notice
- Useful: 3/5
- Signal: 0/5
### Fetched artifact HTML has the server-appended comments island stripped
Reading back a published artifact no longer dumps the server's trailing comments blob into the HTML.
**What**
Reading a published artifact back no longer leaves the server's trailing comments payload embedded in the HTML you get.
**Details**
- A trailing JSON script block holding the comments payload and its MAC is removed before the HTML is used.
- The reported byte count is recomputed from the stripped text rather than the raw response.
- The fetch result now also carries the page favicon.
- The read telemetry event name is parameterized by caller instead of hardcoded.
**Evidence**
`