Feature flags, all 2,095.

tengu_new_init ◇ No description from anybody.never mined ·never mined never read tengu_nifty_finch ◇ Claude Code can now work out an alternate model to suggest from the current one, behind a server switchv2.1.286 → v2.1.309 ◆in the build AND v2.1.286 tengu_nifty_lemur ◇ Five analytics events cover the weekly limit-reset flow.v2.1.248 → v2.1.309 ◆in the build AND v2.1.309 tengu_nimble_wren ◇ The built-in web-fetch helper now uses Haiku when certain model and provider checks pass, and is told not to retry sites that refuse itv2.1.300 → v2.1.309 ◆in the build AND v2.1.302 tengu_noble_mountain ◇ Write tool now accepts file_text, file_content, and path as aliases for content and file_path, rewriting input before plugin hooks see itv2.1.280 → v2.1.309 ◆in the build AND v2.1.280 tengu_noble_snowflake ◇ A new telemetry event, tengu_bridge_transcript_deferral, reports each stage of Remote Control's deferred transcript uploadv2.1.281 → v2.1.309 ◆in the build AND v2.1.281 tengu_noble_wozniak ◇ When the agents view opens, background housekeeping can now wait until you have trusted the folder instead of starting straight awayv2.1.300 → v2.1.309 ◆in the build AND v2.1.300 tengu_non_deferrable_builtins ◇ Default null. A list of tool names that tool search does not defer, so they are loaded with the request. The value may be a plain array or an object keyed by model name substring with a * fallback; it is merged with a non_deferrable_builtins list from another config source, and its entries also feed a set of MCP server names alongside the always-allow rules.v2.1.176 → v2.1.309 ◆in the build AND v2.1.309 tengu_nonblocking_stdout ◇ Failures installing the fast, non-blocking stdout writer now log more detail and report telemetry once per reasonnever mined ·never mined never read tengu_noreread_q7m_velvet ◇ No description from anybody.never mined ·never mined AND v2.1.91 tengu_notification_method_used ◇ Terminal notifications go through one path, and an unrecognised notification setting is reported as invalid.never mined ·never mined never read tengu_oauth_401_recovered_from_disk ◇ After a token refresh, requests use your current account rather than the previous one.never mined ·never mined never read tengu_oauth_401_recovered_from_keychain ◇ An analytics event, not a feature flag. Claude Code logs it while handling an OAuth 401 when the stored credentials, read again, already hold an access token different from the one that was rejected, so that token is used without a refresh.never mined ·never mined never read tengu_oauth_console_api_key_selected ◇ /login now offers signing in to your Console account instead of only creating an API key.never mined ·never mined never read tengu_oauth_console_token_selected ◇ /login now offers signing in to your Console account instead of only creating an API key.never mined ·never mined never read tengu_oauth_error ◇ An analytics event, not a feature flag. Claude Code logs it when an OAuth login flow fails: in the login flow with account_on_hold: true when the account is on hold, otherwise with error details and ssl_error (whether the failure was recognised as an SSL error), and in the install-github-app OAuth flow with error details.never mined ·never mined never read tengu_oauth_refresh_invalid_scope_fallback ◇ If a server rejects the new expanded OAuth scope request, Claude Code retries once with the old scopesnever mined ·never mined never read tengu_oauth_refresh_token_account_on_hold ◇ OAuth token refresh now recognizes an 'account on hold' response as its own outcomenever mined ·never mined never read tengu_oauth_refresh_token_cleared_invalid_grant ◇ No description from anybody.never mined ·never mined never read tengu_oauth_token_refresh_failure ◇ Login token refresh failures are now labelled with the action that triggered them.never mined ·never mined never read tengu_oauth_token_refresh_lock_acquired ◇ OAuth token refresh lock now reports its "acquired" event later, closer to the actual read checknever mined ·never mined never read tengu_oauth_token_refresh_lock_acquiring ◇ OAuth token refresh lock now reports its "acquired" event later, closer to the actual read checknever mined ·never mined never read tengu_oauth_token_refresh_lock_release_error ◇ A telemetry failure can no longer block releasing the OAuth token refresh locknever mined ·never mined never read tengu_oauth_token_refresh_lock_releasing ◇ A telemetry failure can no longer block releasing the OAuth token refresh locknever mined ·never mined never read tengu_oauth_token_refresh_locked_read_failed ◇ Login token refresh retries saving and stops cleanly on dead tokens instead of failing quietly.never mined ·never mined never read tengu_oauth_token_refresh_race_resolved ◇ When another process refreshes your login, the session re-reads who you are signed in as.never mined ·never mined never read tengu_oauth_token_refresh_starting ◇ An analytics event, not a feature flag. Claude Code logs it, with no fields, just before it sends an OAuth token refresh request, after it has checked that a refresh token exists, the token still needs refreshing, the refresh token is not known dead or on hold, and the refresh lock is not compromised.never mined ·never mined never read tengu_oauth_token_refresh_success ◇ Token refresh now also asks for the user:plugins permission scope and logs whether it was grantednever mined ·never mined never read tengu_oauth_tokens_save_exception ◇ Credential storage telemetry now reports which backend or fallback chain was actually usednever mined ·never mined never read tengu_oauth_tokens_save_failed ◇ Credential storage telemetry now reports which backend or fallback chain was actually usednever mined ·never mined never read tengu_oauth_tokens_saved ◇ Saved login tokens now propagate into the newer storage-backed credential cache.never mined ·never mined never read tengu_oboe ◇ No description from anybody.never mined ·never mined never read tengu_observer_agents_enabled ◇ Default true. Kill switch for observer agents, the observer field in an agent definition: they also require CLAUDE_CODE_EXPERIMENTAL_OBSERVER_AGENTS, and a false value here leaves every agent unobserved. The same gate also switches a second code path elsewhere in the bundle.v2.1.200 → v2.1.309 ◆in the build AND v2.1.200 tengu_observer_subagent_fanout ◇ Default true. When false, an observer pairing is not extended to the subagents an observed agent spawns. When on, fan-out still requires observer agents to be enabled, stops at depth 2, and is skipped for agents with observeSubagents: false.v2.1.218 → v2.1.309 ◆in the build AND v2.1.218 tengu_ochre_bittern ◇ A helper that always returned false now checks a remote gate first.v2.1.260 → v2.1.309 ◆in the build AND v2.1.260 tengu_ochre_dunlin ◇ When staged files fail to land, Claude Code logs a warning and can tell the model the files may be missing or out of datev2.1.295 → v2.1.309 ◆in the build AND v2.1.300 tengu_ochre_finch ◇ Default false. When true, the memory prompt's types section becomes a short list of the four memory types (user, feedback, project, reference) with one-line descriptions and an instruction to invoke the memory-types skill, and that skill (not user-invocable) is enabled when auto memory is on.v2.1.138 or earlier → v2.1.309 ◆in the build never read tengu_ochre_gantry ◇ Default {}. Configures the Large workflow warning on dynamic workflow rows: enabled: false turns it off, agents sets the scheduled agent cap (default 25) and tokens the token cap (default 1500000). The warning shows when scheduled agents exceed the cap or actual or projected tokens exceed the token cap; CLAUDE_CODE_WORKFLOW_SIZE_WARNING_AGENTS and the workflowSizeGuideline setting take precedence for agents, CLAUDE_CODE_WORKFLOW_SIZE_WARNING_TOKENS for tokens, and no warning is shown while ultracode is active.v2.1.203 → v2.1.309 ◆in the build never read tengu_ochre_hollow ◇ Artifact syntax highlighting and mermaid diagrams are now always on; three other gates were renamed and stay off.v2.1.138 or earlier → v2.1.238 ○gone AND v2.1.239 tengu_ochre_kite ◇ Claude Code now recovers gracefully when a safety filter cuts off a response mid-tool-callv2.1.273 → v2.1.281 ○gone AND v2.1.282 tengu_ochre_plover ◇ Claude can auto-draft replies to artifact comments, but it's off by default and declines by default.v2.1.260 → v2.1.309 ◆in the build AND v2.1.260 tengu_ochre_wren ◇ An alternate opening line for Claude's system prompt can be switched on with an environment variable.v2.1.239 → v2.1.294 ○gone AND v2.1.248 tengu_official_plugin_prompt_overrides ◇ Default {}. Map from plugin name to override text for plugins from an official marketplace (or another allowed plugin set). Each entry may carry server_instructions, server_instructions_by_server, tools, search_hints, param_descriptions, prompts and skills; a non-object payload is ignored, and an entry that fails the schema is logged and the baked-in text is used.v2.1.208 → v2.1.309 ◆in the build never read tengu_omelette_fouet ◇ A startup tip pointing at a /design command for mocking up UI is present but nobody sees it.v2.1.199 → v2.1.281 ○gone never read tengu_omelette_grant_watch ◇ Default false. When true, while Claude Code is asking you to approve the first write under a Claude Design project grant, it polls /v1/design/grants and, when a grant for that project newer than the one it first saw appears (for example one approved on claude.ai), records it so the pending prompt can resolve without an answer in the terminal. When off, no such watcher runs.v2.1.213 → v2.1.309 ◆in the build never read tengu_omelette_whisk ◇ Default false. On the first-party API provider, adds the built-in claude_design MCP server (HTTP, https://api.anthropic.com/v1/design/mcp) to the dynamic MCP servers. The env var CLAUDE_CODE_ENABLE_DESIGN_MCP takes precedence over this flag when it is set.never mined ·never mined never read tengu_on_branch_default_guard_observe ◇ Cloud on-branch checkout guard can now be forced on via a CCR_ON_BRANCH_DEFAULT_GUARD environment variablev2.1.227 → v2.1.241 ○gone never read tengu_onboarding_step ◇ Onboarding now also treats a 'gateway' authentication setup as equivalent to OAuth for its flownever mined ·never mined never read tengu_onyx_basin_m1k ◇ No description from anybody.never mined ·never mined never read tengu_onyx_plover ◇ Default null. Configures auto dream memory consolidation: enabled: true turns it on when the autoDreamEnabled setting is unset, enabled or available being true makes the feature available, and minHours (default 24) and minSessions (default 5) set how long and how many sessions must pass since the last consolidation before another runs. Non-positive or non-numeric values use the defaults.v2.1.138 or earlier → v2.1.309 ◆in the build AND v2.1.94 tengu_onyx_sluice ◇ Artifact comments could be fetched server-side instead of through the frame host, but that switch is off.v2.1.228 → v2.1.309 ◆in the build AND v2.1.281 tengu_opal_circuit ◇ The nudge suggesting the verify skill is now silenced unless a gate turns it on.never mined ·never mined never read tengu_open_agents_via_left ◇ Sending a session to a background agent succeeds when the worker is already running.never mined ·never mined never read tengu_opus_default_pro_plan ◇ No description from anybody.never mined ·never mined never read tengu_opus_to_opus1m_migration ◇ An analytics event, not a feature flag. Claude Code logs it, with no fields, when a startup migration rewrites a user settings model of opus to opus[1m] and the write succeeds. The migration does nothing when the user setting is not opus or when an eligibility check fails (for example on a non-first-party API provider).never mined ·never mined never read tengu_orchid_mantis ◇ Default false. When on and tengu_orchid_mantis_v2 is off, the system prompt's session guidance tells the model it may end a reply with a one-line offer to /schedule a background agent for a natural future follow-up, with signals for one-time and recurring offers; this needs the Skill tool and a schedule skill to be available. Separately, it switches the /schedule skill's when-to-use text to one that also covers one-time runs such as "open a cleanup PR for X in 2 weeks".v2.1.138 or earlier → v2.1.205 ○gone AND v2.1.272 tengu_orchid_mantis_v2 ◇ Default false. When on, and the Skill tool and a schedule skill are available, the system prompt's session guidance carries a stricter /schedule rule: no offer by default, and an offer only when the turn left a named artifact with a quotable future obligation (a flag with a ramp date, a .skip with a removal condition, a job ID with an ETA, a dated TODO), at most once per session. In that guidance it takes precedence over tengu_orchid_mantis.v2.1.138 or earlier → v2.1.205 ○gone AND v2.1.272 tengu_orderly_lantern ◇ WebFetch URL approval prompts now remember approved addresses and can retry without asking, and a switch can make them appear one at a timev2.1.292 → v2.1.309 ◆in the build AND v2.1.292 tengu_orford_ness ◇ Default false. Turns on sending error reports (logged errors, unhandled rejections and uncaught exceptions) to Datadog error tracking, with a per-process report cap. It is checked last: reporting also needs DISABLE_ERROR_REPORTING unset, the first-party API provider, a Claude Code version of at least 2.1.193, the allow_error_reporting policy check passing, and other internal checks.v2.1.191 → v2.1.274 ○gone AND v2.1.309 tengu_org_config_gate_shadow ◇ Claude Code now reports why fetching organization settings succeeded or failed, and logs what an organization settings check would decidev2.1.295 ○gone never read tengu_org_config_required ◇ No description from anybody.v2.1.293 → v2.1.309 ◆in the build AND v2.1.309 tengu_org_memory_connected_mode ◇ Claude records whether connected org memory is in use when you haven't listed stores yourself.never mined ·never mined never read tengu_org_memory_decision ◇ Org memory remembers who enabled it, and a different account trying to enable it is rejected.never mined ·never mined never read tengu_org_memory_decision_reopened ◇ Shared org memory now remembers who settled a decision and lets a settled decision be reopened.never mined ·never mined never read tengu_org_memory_first_write_into_write_home ◇ Team memory sync can now skip an individual refused file without breaking the whole sync, and reports how many files were refusednever mined ·never mined never read tengu_org_memory_legacy_pick_nudged ◇ The memory store picker nudges you to pick a specific project instead of leaving the legacy default.never mined ·never mined never read tengu_org_memory_project_selected ◇ The org memory project picker now says whether your switch applies now or next session.never mined ·never mined never read tengu_org_memory_project_switch ◇ Switching shared project memory now reports exactly what happened and marks old results stale.never mined ·never mined never read tengu_org_memory_root_mount_derived ◇ Org memory setup renames aside anything squatting its mount name rather than colliding.never mined ·never mined never read tengu_org_memory_root_write_outcome ◇ Org memory now falls back to read-only when write permission is ambiguous, with new logging for why.never mined ·never mined never read tengu_org_memory_token_discarded ◇ Org memory now falls back to read-only when write permission is ambiguous, with new logging for why.never mined ·never mined never read tengu_org_memory_write_opt_in_requested ◇ Claude logs an event just before asking you for an org-memory write credential.never mined ·never mined never read tengu_org_memory_writes_monorepo_blocked ◇ No description from anybody.never mined ·never mined never read tengu_org_penguin_mode_fetch_failed ◇ Fast mode ('penguin mode') gains an admin setting to require per-session opt-in or disable it, plus an org-check bypass env varnever mined ·never mined never read tengu_org_penguin_mode_fetch_refused ◇ Fast mode ('penguin mode') gains an admin setting to require per-session opt-in or disable it, plus an org-check bypass env varnever mined ·never mined never read tengu_org_policy_denied ◇ Blocked remote-session actions from org policy now emit a tengu_org_policy_denied telemetry eventnever mined ·never mined AND v2.1.296 tengu_orphaned_permission_applied ◇ Three telemetry events now track permissions parked across turn boundaries and whether they were applied.never mined ·never mined never read tengu_osier_pylon_trace ◇ Claude could read console errors and failed loads from real viewers of an artifact, but it's switched off.v2.1.238 → v2.1.309 ◆in the build AND v2.1.280 tengu_outputs_push_completed ◇ Cloud sessions now upload files you changed during a turn back to the session worker.never mined ·never mined never read tengu_pale_kestrel ◇ Permission prompts for inline shell commands such as bash -c now depend on whether the command contains rm or rmdirv2.1.299 → v2.1.309 ◆in the build AND v2.1.299 tengu_paper_halyard ◇ Internal rewrite of how instruction files get filtered from context, no visible change.v2.1.138 or earlier → v2.1.309 ◆in the build AND v2.1.252 tengu_parallel_trinket ◇ Bash tool on Windows can now launch commands through a dedicated PowerShell/cmd launcherv2.1.269 → v2.1.309 ◆in the build never read tengu_parchment_fern ◇ An alternate prompt wording for two tools exists but stays off unless you set an environment variable.v2.1.248 → v2.1.283 ○gone AND v2.1.223 tengu_parsed_eagle ◇ Claude Code can re-encode images as WebP when that is smaller, and now refuses images that are too costly to decodev2.1.295 → v2.1.309 ◆in the build AND v2.1.295 tengu_parsed_willow ◇ New tengu_parsed_willow gate (defaulting on) added with CLAUDE_CODE_PARSED_WILLOW env overridev2.1.277 → v2.1.309 ◆in the build AND v2.1.281 tengu_partial_stream_retraction_closed ◇ The code that withdraws a half-printed reply and shows a refusal banner was reorganized; nothing looks different.never mined ·never mined never read tengu_passport_quail ◇ Memory extraction can now turn itself on when post-turn-memory config is present, bypassing its usual feature gatev2.1.138 or earlier → v2.1.309 ◆in the build AND v2.1.305 tengu_patient_finch ◇ If a URL failed earlier in the session, WebFetch skips a quick second attempt and tells the model to say the page could not be readv2.1.300 → v2.1.309 ◆in the build AND v2.1.300 tengu_patient_meteor ◇ When a subagent edits a file, the Edit tool can attach a compact summary result, off unless switched on remotelyv2.1.295 → v2.1.309 ◆in the build AND v2.1.295 tengu_patient_tide ◇ Claude Code can now get an extra retry allowance from the server for main-conversation requests, unless you set a retry variablev2.1.309 ●new AND v2.1.309 tengu_pdf_page_extraction ◇ PDF extraction and the retention sweep now receive a storage handle instead of finding one themselves.never mined ·never mined never read tengu_pdf_reference_attachment ◇ PDF page-count detection now reports specific failure reasons and handles PDFs with an unknown page countnever mined ·never mined never read tengu_peaceful_leaf ◇ A remote switch can add guidance on files and folders on your computer to the browser and computer-use tool descriptionsv2.1.288 → v2.1.309 ◆in the build AND v2.1.299 tengu_pencil_farmer ◇ Default false. Picks the queue priority of a message recognized as a verified Slack human turn when nothing else sets it: next when on, later when off. It stays later when another such Slack turn at later priority is already queued or running, and other checks can also force later.v2.1.193 → v2.1.309 ◆in the build never read tengu_penguins_enabled ◇ No description from anybody.never mined ·never mined AND v2.1.36 tengu_penguins_off ◇ Fast mode ('penguin mode') gains an admin setting to require per-session opt-in or disable it, plus an org-check bypass env varv2.1.138 or earlier → v2.1.309 ◆in the build AND v2.1.299 tengu_peppy_riddle ◇ A check that matched MCP servers against a remote list can now read a single on/off switch instead, depending on how it is calledv2.1.300 → v2.1.304 ○gone AND v2.1.300 tengu_peppy_zephyr ◇ Tool token counts for context analysis can come from the server instead of local estimates, but the gate is off.v2.1.257 → v2.1.276 ○gone AND v2.1.261 tengu_per_turn_effort ◇ Per-family capability gates now use a name-to-flag table instead of matching skill and tool name strings.never mined ·never mined AND v2.1.309 tengu_permission_explainer ◇ No description from anybody.never mined ·never mined never read