alysis-code 0.13.7__tar.gz → 0.14.0__tar.gz

{alysis_code-0.13.7 → alysis_code-0.14.0}/.gitignore

@@ -14,6 +14,7 @@ venv/
 
 # Tooling
 .dev/
+.worktrees/
 .pytest_cache/
 .pytest-*/
 .ruff_cache/
@@ -45,6 +46,10 @@ extensions/*/out/
 extensions/*/*.vsix
 
 # Local review/context bundles
+# Retired internal report directories, regardless of artifact format.
+/docs/internal/
+/docs/subagents/
+/docs/tooling/
 *_context_review.zip
 owner-scoping.patch
 alysis_ecs.tar.gz

{alysis_code-0.13.7 → alysis_code-0.14.0}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: alysis-code
-Version: 0.13.7
-Summary: Open-source coding agent for hosted and local models.
+Version: 0.14.0
+Summary: Local CLI coding agent powered by any API.
 Project-URL: Changelog, https://github.com/AlysisAi/alysis-code/blob/main/docs/CHANGELOG.md
 Project-URL: Documentation, https://github.com/AlysisAi/alysis-code/tree/main/docs
 Project-URL: Funding, https://github.com/sponsors/AlysisAi
@@ -25,17 +25,19 @@ Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Classifier: Topic :: Software Development
 Classifier: Topic :: Utilities
 Requires-Python: >=3.11
+Requires-Dist: anyio>=4.14.2
 Requires-Dist: click>=8.0.0
 Requires-Dist: cryptography>=50.0.0
 Requires-Dist: ddgs>=9.0.0
-Requires-Dist: h2>=4.4.1
 Requires-Dist: httpcore<2.0.0,>=1.0.0
 Requires-Dist: httpx>=0.27.0
 Requires-Dist: keyring>=25.0.0
+Requires-Dist: markdown-it-py<5.0.0,>=4.2.0
 Requires-Dist: packaging>=23.0
 Requires-Dist: pillow>=12.3.0
 Requires-Dist: platformdirs>=4.2.0
 Requires-Dist: prompt-toolkit>=3.0.43
+Requires-Dist: psutil<8.0,>=7.2.2
 Requires-Dist: pydantic>=2.6.0
 Requires-Dist: pygments<3.0.0,>=2.17.0
 Requires-Dist: rich>=13.7.1
@@ -148,16 +150,18 @@ Shell and verification commands default to strict sandboxing. Verification can r
 sandbox only when explicitly configured with `verify_sandbox.mode="off"` or
 `ALYSIS_VERIFY_SANDBOX_MODE=off`.
 
-## Modes & Personas
+## Permissions & Personas
 
 Two layers control what the agent may do — one enforces, one focuses.
 
-**Execution modes** are the enforcement layer. Choose one per command with `--mode`, switch in chat
-with `/mode`, or set a default with `alysis-code config set default_mode <mode>`:
+**Permissions** are the enforcement layer. Choose one per command with `--mode`, select the next
+message's level in chat with `/permissions` or **Shift+Tab**, or set a default with
+`alysis-code config set default_mode <mode>`. A selection takes effect when the next user message
+starts; it does not change a turn already running:
 
 | Mode | Behavior |
 | --- | --- |
-| `readonly` | Inspection only — no file writes, shell, MCP, or subagent delegation. |
+| `readonly` | Inspection plus top-level read-only delegation — no file writes, shell, MCP, candidate apply/discard, or nested delegation. |
 | `review` | The safe default — previews and asks before every write and shell command. |
 | `auto` | Applies routine edits with fewer prompts; dangerous operations stay blocked. |
 | `fullaccess` | No mode-level prompts for trusted workspaces — the denylist and audit log stay active. |
@@ -176,6 +180,10 @@ A persona is a convention, never a permission: it can keep or lower your executi
 never raise it. Define your own in `.alysis_personas/*.md`.
 [Learn more](https://alysiscode.com/docs/concepts/personas)
 
+Use `/persona architect` for planning in chat, or `/forge` for a plan-driven execution workflow.
+The former chat `/plan` commands have been removed; Forge retains its plan views and editor.
+Persona changes made during a running turn apply after that turn finishes.
+
 ## Project Links
 
 - [Website](https://alysiscode.com/)
@@ -209,4 +217,4 @@ More questions? See the [full FAQ](https://alysiscode.com/docs/faq).
 
 ---
 
-**Join the community** — [X](https://x.com/alysiscode) | [GitHub Issues](../../issues)
+**Join the community** — [X](https://x.com/alysiscode) | [GitHub Issues](https://github.com/AlysisAi/alysis-code/issues)

{alysis_code-0.13.7 → alysis_code-0.14.0}/README.md

@@ -89,16 +89,18 @@ Shell and verification commands default to strict sandboxing. Verification can r
 sandbox only when explicitly configured with `verify_sandbox.mode="off"` or
 `ALYSIS_VERIFY_SANDBOX_MODE=off`.
 
-## Modes & Personas
+## Permissions & Personas
 
 Two layers control what the agent may do — one enforces, one focuses.
 
-**Execution modes** are the enforcement layer. Choose one per command with `--mode`, switch in chat
-with `/mode`, or set a default with `alysis-code config set default_mode <mode>`:
+**Permissions** are the enforcement layer. Choose one per command with `--mode`, select the next
+message's level in chat with `/permissions` or **Shift+Tab**, or set a default with
+`alysis-code config set default_mode <mode>`. A selection takes effect when the next user message
+starts; it does not change a turn already running:
 
 | Mode | Behavior |
 | --- | --- |
-| `readonly` | Inspection only — no file writes, shell, MCP, or subagent delegation. |
+| `readonly` | Inspection plus top-level read-only delegation — no file writes, shell, MCP, candidate apply/discard, or nested delegation. |
 | `review` | The safe default — previews and asks before every write and shell command. |
 | `auto` | Applies routine edits with fewer prompts; dangerous operations stay blocked. |
 | `fullaccess` | No mode-level prompts for trusted workspaces — the denylist and audit log stay active. |
@@ -117,6 +119,10 @@ A persona is a convention, never a permission: it can keep or lower your executi
 never raise it. Define your own in `.alysis_personas/*.md`.
 [Learn more](https://alysiscode.com/docs/concepts/personas)
 
+Use `/persona architect` for planning in chat, or `/forge` for a plan-driven execution workflow.
+The former chat `/plan` commands have been removed; Forge retains its plan views and editor.
+Persona changes made during a running turn apply after that turn finishes.
+
 ## Project Links
 
 - [Website](https://alysiscode.com/)
@@ -150,4 +156,4 @@ More questions? See the [full FAQ](https://alysiscode.com/docs/faq).
 
 ---
 
-**Join the community** — [X](https://x.com/alysiscode) | [GitHub Issues](../../issues)
+**Join the community** — [X](https://x.com/alysiscode) | [GitHub Issues](https://github.com/AlysisAi/alysis-code/issues)

{alysis_code-0.13.7 → alysis_code-0.14.0}/docs/CHANGELOG.md

@@ -8,6 +8,77 @@ Notable user-facing changes to Alysis Code are recorded here. This project follo
 
 No user-facing changes yet.
 
+## [0.14.0] - 2026-09-24
+
+### Added
+
+- Add GPT-6 Sol and GPT-6 Luna to the OpenAI API and ChatGPT subscription model
+  choices, with pricing, context limits, and reasoning settings. Subscription
+  availability follows the live account catalog.
+- Add Claude Opus 5.5 to Anthropic and OpenRouter, Grok 4.7 to xAI and OpenRouter,
+  Mistral's hosted GLM 5.3, and current OpenRouter model routes.
+- Show DeepSeek V4.1 Flash and GLM 5.3 Flash as hosted account model choices.
+- Add `/objective` commands to inspect, replace, and amend the current task across
+  retries and resumed sessions.
+- Allow a configurable set of public package registries for `web_fetch` while
+  retaining URL and network safety checks.
+
+### Changed
+
+- Show clearer model names in setup and move the request timeout control to
+  Advanced settings.
+- Improve subagent scheduling, task evidence, and verification reporting.
+
+### Fixed
+
+- Suppress unexecuted tool-call markup in replies, retry once with structured
+  tool calls, and report a failure if the model repeats the malformed output.
+- Require AnyIO 4.14.2 or newer in the HTTP runtime dependency set.
+- Keep hosted account credentials on the selected gateway when an unsupported
+  model is chosen, and preserve retired Xiaomi selections without silently
+  redirecting them to a different provider's model.
+- Omit unsupported temperature settings for GPT-6 reasoning requests and reject
+  attempts to disable reasoning for always-thinking OpenRouter models.
+- Improve session recovery and deadline handling for long running work.
+- Verify durable service listener ownership on macOS without requiring root access.
+
+## [0.13.8] - 2026-09-16
+
+### Changed
+
+- Remove legacy chat Plan Mode, its draft/approve flow, and its prompts and toolbar
+  state. Use the Architect persona for planning in chat; Forge retains its plan views
+  and editor.
+- Replace `/mode` with `/permissions`. Permission selections and Shift+Tab apply to
+  the next user message, with the pending choice visible beside the active permissions.
+- Apply persona changes and approved persona-switch proposals only after the active
+  turn finishes. Preserve pending permission selections when resuming a conversation.
+- Show pending steering messages and queued follow-ups. Shift+Left recalls the newest
+  queued message for editing. Interruptions preserve undelivered work and wait for the
+  current worker to finish before another turn starts.
+- Read large Git diffs in complete pages with a content identifier that detects
+  changes between pages. The agent can select a path or inspect staged changes.
+- Allow a background-process wait of up to 900 seconds, bounded by the remaining
+  run budget and returning early when its requested condition is met.
+- Explain how to correct rejected verification commands, including how to use the
+  session's configured verification contract.
+
+### Fixed
+
+- Keep plain assistant replies in conversation history so follow-up requests retain
+  the previous answer even when the provider returns no extra metadata.
+- Redact credential fields inside nested JSON, tool arguments, feedback text, and
+  support bundles. Omit data that exceeds the safe redaction depth and non-UTF-8
+  diagnostic artifacts that cannot be sanitized reliably.
+- Clean up incomplete support bundles after cancellation or I/O errors, support long
+  Windows export paths, and open report drafts without disrupting terminal output.
+- Inspect Git changes without running configured diff programs, text converters,
+  filesystem monitors, or content filters. Worktree reads that require content
+  filters or submodule helpers return guidance for an approved shell command;
+  staged diffs and unaffected scoped paths remain available.
+- Preserve hosted-account explanations for rolling usage limits, concurrent requests,
+  reserved credits, and temporary routing or capacity problems.
+
 ## [0.13.7] - 2026-09-11
 
 ### Added
@@ -271,7 +342,9 @@ Public launch and repository-layout refresh.
 Release notes for versions before 0.9.8 remain available in the
 [GitHub Releases archive](https://github.com/AlysisAi/alysis-code/releases).
 
-[Unreleased]: https://github.com/AlysisAi/alysis-code/compare/v0.13.7...HEAD
+[Unreleased]: https://github.com/AlysisAi/alysis-code/compare/v0.14.0...HEAD
+[0.14.0]: https://github.com/AlysisAi/alysis-code/compare/v0.13.8...v0.14.0
+[0.13.8]: https://github.com/AlysisAi/alysis-code/compare/v0.13.7...v0.13.8
 [0.13.7]: https://github.com/AlysisAi/alysis-code/compare/v0.13.6.2...v0.13.7
 [0.13.6.2]: https://github.com/AlysisAi/alysis-code/compare/v0.13.6.1...v0.13.6.2
 [0.13.6.1]: https://github.com/AlysisAi/alysis-code/compare/v0.13.6...v0.13.6.1

{alysis_code-0.13.7 → alysis_code-0.14.0}/pyproject.toml

@@ -4,8 +4,8 @@ build-backend = "hatchling.build"
 
 [project]
 name = "alysis-code"
-version = "0.13.7"
-description = "Open-source coding agent for hosted and local models."
+version = "0.14.0"
+description = "Local CLI coding agent powered by any API."
 readme = "README.md"
 requires-python = ">=3.11"
 license = { text = "Apache-2.0" }
@@ -32,18 +32,20 @@ classifiers = [
   "Topic :: Utilities",
 ]
 dependencies = [
+  "anyio>=4.14.2",
   "click>=8.0.0",
   "cryptography>=50.0.0",
   "ddgs>=9.0.0",
-  "h2>=4.4.1",
   "httpcore>=1.0.0,<2.0.0",
   "httpx>=0.27.0",
   "keyring>=25.0.0",
+  "markdown-it-py>=4.2.0,<5.0.0",
   "packaging>=23.0",
   "pillow>=12.3.0",
   "platformdirs>=4.2.0",
   "pydantic>=2.6.0",
   "prompt-toolkit>=3.0.43",
+  "psutil>=7.2.2,<8.0",
   "pygments>=2.17.0,<3.0.0",
   "rich>=13.7.1",
   "tiktoken>=0.7.0",

{alysis_code-0.13.7 → alysis_code-0.14.0}/src/alysis_code/__init__.py

@@ -1,3 +1,3 @@
 __all__ = ["__version__"]
 
-__version__ = "0.13.7"
+__version__ = "0.14.0"

{alysis_code-0.13.7 → alysis_code-0.14.0}/src/alysis_code/_build_info.py

@@ -8,7 +8,9 @@ stale commit hash.
 """
 
 BUILD_COMMIT = ""
+BUILD_COMMIT_SUBJECT = ""
 BUILD_TIMESTAMP = ""
 BUILD_DIRTY = True
 BUILD_SOURCE = "dev-default"
-BUILD_INFO_SCHEMA_VERSION = 1
+BUILD_RESOLVED_ORIGIN_MAIN = "unavailable"
+BUILD_INFO_SCHEMA_VERSION = 2

{alysis_code-0.13.7 → alysis_code-0.14.0}/src/alysis_code/account_login.py

@@ -37,7 +37,7 @@ from .config import (
     save_persisted_profile_key,
 )
 from .host_browser import open_url
-from .profile_presets import get_preset, make_profile_from_preset
+from .profile_presets import canonical_model_alias_for_preset, get_preset, make_profile_from_preset
 from .profiles import ProfileSpec, add_profile, get_profile, set_active_profile
 
 # Matches the server-side device-code expiry (15 min) so we poll until the code
@@ -49,7 +49,7 @@ _DEFAULT_POLL_INTERVAL_S = 5.0
 # model picker, so an offline/slow gateway must not stall the menu for long.
 _MODELS_TIMEOUT_S = 6.0
 
-_FRESH_LOGIN_DEFAULT_MODEL = "deepseek-v4-flash"
+_FRESH_LOGIN_DEFAULT_MODEL = "deepseek-flash"
 
 
 class AlysisLoginError(Exception):
@@ -441,11 +441,11 @@ def _format_token_usage(used: int | None, total: int | None) -> str | None:
 def _activate_alysis_profile(cfg: AppConfig, *, email: str | None) -> LoginResult:
     """Create + activate the `alysis` profile, keeping any model the user chose.
 
-    Fresh logins default to the Pro flagship (``deepseek-v4-flash``) so
-    subscribe → login → chat works with zero extra steps. Re-logins preserve
+    Fresh logins default to V4.1 Flash (``deepseek-flash``) so
+    connect → login → chat works with zero extra steps. Re-logins preserve
     whatever model the user selected since, so logging in again never undoes
-    their choice. Legacy MiMo ids (the retired Xiaomi trial) are remapped to
-    the current default via the preset's model_aliases at config load.
+    supported choices. Retired Flash/beta ids are remapped to the current
+    default via the preset's model_aliases, including on re-login.
     """
     preset = get_preset(cloud.PROFILE_KEY)
     if preset is not None:
@@ -456,6 +456,8 @@ def _activate_alysis_profile(cfg: AppConfig, *, email: str | None) -> LoginResul
     existing = get_profile(cfg, cloud.PROFILE_KEY)
     existing_model = str(getattr(existing, "default_model", "") or "").strip() if existing else ""
     chosen_model = existing_model or _FRESH_LOGIN_DEFAULT_MODEL
+    if preset is not None:
+        chosen_model = canonical_model_alias_for_preset(preset, chosen_model)
 
     # Always pin to the live gateway URL (env-overridable for tests); keep the
     # user's chosen model.

{alysis_code-0.13.7 → alysis_code-0.14.0}/src/alysis_code/agent/README.md

@@ -12,6 +12,11 @@ managed execution flows.
 - [turn_path.py](turn_path.py) defines execution posture and reply-language directives.
 - [prompt_context.py](prompt_context.py) prepares workspace, convention, skill, plugin, and
   verification context.
+- [task_state.py](task_state.py) holds the host-owned task identity (`SessionTaskState`):
+  the accepted objective, its originating user event, and its transitions.
+  The pinned `<task_brief>` is a bounded rendering of it, installed before the
+  first model request of a turn; it is never inferred from edits, tool output,
+  or model text, and it grants no permission.
 - [verification.py](verification.py) and [verification_commands.py](verification_commands.py)
   support verification flows.