aepkit 0.7.3__tar.gz → 0.7.4__tar.gz

{aepkit-0.7.3 → aepkit-0.7.4}/CHANGELOG.md

@@ -7,6 +7,30 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.7.4] - 2026-11-08
+
+### Added
+- `references/autonomous-mode.md`: "Tool versions at the start of a run". Before the first task
+  the Auditor updates the tools the run depends on (`aepkit`, any plugin or harness) and records
+  their versions in the plan of record; the Auditor brief gains a `TOOLS` field for it.
+- `aeprogress` 0.4.0:
+  - Expandable phases: each phase row opens its full title, its bar and count, and its tasks and
+    gate as state chips (approved, conditional, in progress with the worker, awaiting audit,
+    pending, rejected). The active phase opens by itself; a phase the owner closes stays closed.
+    Where the surface reports hover (desktop app, fullscreen terminal), a closed phase shows the
+    same in a card.
+  - Update check: the pane and `/aeprogress` say when a newer version exists, from the
+    marketplace catalog Claude Code keeps on disk and, at most once a day, from the version
+    published on GitHub (one request for a public file; option **Update check**: `daily`,
+    `offline` or `off`; skipped for copies from the claude.ai directory and when
+    `CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC` is set). The README explains how to turn on
+    auto-update for this marketplace.
+  - An icon (`.claude-plugin/icon.png`, source in `assets/icon.svg`), and `documentationUrl`,
+    `supportUrl` and `privacyPolicyUrl` in `plugin.json`, for the plugin directory. The README
+    gains "What it does on your machine" (every process it starts, the files and variables it
+    reads, the one network request, and what it can refuse) and "Privacy", and shows its
+    screenshots as Markdown images inside the plugin.
+
 ## [0.7.3] - 2026-11-08
 
 No change to the package's code. This release puts the new README on PyPI: the protocol's
@@ -369,7 +393,8 @@ Fixes from an independent review of 0.3.0.
 - Multi-framework agent installation support for Antigravity, Claude Code, and Cursor.
 - Zero-dependency test runner (`tests/run.py`).
 
-[Unreleased]: https://github.com/tBeltty/auditor-executor-protocol/compare/v0.7.3...HEAD
+[Unreleased]: https://github.com/tBeltty/auditor-executor-protocol/compare/v0.7.4...HEAD
+[0.7.4]: https://github.com/tBeltty/auditor-executor-protocol/compare/v0.7.3...v0.7.4
 [0.7.3]: https://github.com/tBeltty/auditor-executor-protocol/compare/v0.7.2...v0.7.3
 [0.7.2]: https://github.com/tBeltty/auditor-executor-protocol/compare/v0.7.1...v0.7.2
 [0.7.1]: https://github.com/tBeltty/auditor-executor-protocol/compare/v0.7.0...v0.7.1

{aepkit-0.7.3/src/aepkit.egg-info → aepkit-0.7.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: aepkit
-Version: 0.7.3
+Version: 0.7.4
 Summary: Mechanical support for the Auditor/Executor protocol: scaffold, lint, and negative-control the document set for a multi-phase, multi-agent execution.
 Author-email: tBeltty <jhonatan@tbelt.online>
 License-Expression: MIT
@@ -183,7 +183,7 @@ A failure only proves something if it is the failure you meant. A syntax error,
 
 `aeprogress` is a Claude Code mod for runs where the Auditor launches Executors as subagents or as workers of a harness such as Orca. The protocol does not depend on it, and it reads the run only through `aepkit status --json`.
 
-<img src="https://raw.githubusercontent.com/tBeltty/auditor-executor-protocol/main/integrations/claude-code/aeprogress/assets/pane.png" alt="The aeprogress pane: 43.3% overall, a bar per phase, P2-T3 in progress on a subagent, P2-T4 awaiting audit, P2-T2 conditional, and the machine at 31% with 1 of 2 workers live" width="420">
+<img src="https://raw.githubusercontent.com/tBeltty/auditor-executor-protocol/main/integrations/claude-code/aeprogress/assets/pane.png" alt="The aeprogress pane: 43.3% overall, a bar per phase, the active phase P2 open with its task chips, P2-T3 in progress on a subagent, P2-T4 awaiting audit and P2-T2 conditional" width="420">
 
 It does three things:
 
@@ -203,7 +203,7 @@ Install it at the prompt of a Claude Code terminal session, then open it:
 /aeprogress docs/<task-name>
 ```
 
-It works in the terminal and in the desktop app's local sessions (in the app: **+** next to the prompt, **Plugins**, **Add marketplace**). Every limit and feature is an option in `/config`. Claude Code marks its mod API as early access, so a Claude Code release can break it. Setup, options and how each number is computed: [integrations/claude-code/aeprogress](https://github.com/tBeltty/auditor-executor-protocol/tree/main/integrations/claude-code/aeprogress).
+It works in the terminal and in the desktop app's local sessions (in the app: **+** next to the prompt, **Plugins**, **Add marketplace**). Every limit and feature is an option in `/config`. It needs Claude Code 2.1.287 or later and sends no data about you anywhere (its one network request, a daily version check, can be turned off); its README lists every process it starts and everything it can refuse. To get new versions on their own, turn on auto-update once: `/plugin`, **Marketplaces**, `auditor-executor-protocol`, **Enable auto-update**. Setup, options and how each number is computed: [integrations/claude-code/aeprogress](https://github.com/tBeltty/auditor-executor-protocol/tree/main/integrations/claude-code/aeprogress).
 
 ---
 

{aepkit-0.7.3 → aepkit-0.7.4}/README.md

@@ -145,7 +145,7 @@ A failure only proves something if it is the failure you meant. A syntax error,
 
 `aeprogress` is a Claude Code mod for runs where the Auditor launches Executors as subagents or as workers of a harness such as Orca. The protocol does not depend on it, and it reads the run only through `aepkit status --json`.
 
-<img src="https://raw.githubusercontent.com/tBeltty/auditor-executor-protocol/main/integrations/claude-code/aeprogress/assets/pane.png" alt="The aeprogress pane: 43.3% overall, a bar per phase, P2-T3 in progress on a subagent, P2-T4 awaiting audit, P2-T2 conditional, and the machine at 31% with 1 of 2 workers live" width="420">
+<img src="https://raw.githubusercontent.com/tBeltty/auditor-executor-protocol/main/integrations/claude-code/aeprogress/assets/pane.png" alt="The aeprogress pane: 43.3% overall, a bar per phase, the active phase P2 open with its task chips, P2-T3 in progress on a subagent, P2-T4 awaiting audit and P2-T2 conditional" width="420">
 
 It does three things:
 
@@ -165,7 +165,7 @@ Install it at the prompt of a Claude Code terminal session, then open it:
 /aeprogress docs/<task-name>
 ```
 
-It works in the terminal and in the desktop app's local sessions (in the app: **+** next to the prompt, **Plugins**, **Add marketplace**). Every limit and feature is an option in `/config`. Claude Code marks its mod API as early access, so a Claude Code release can break it. Setup, options and how each number is computed: [integrations/claude-code/aeprogress](https://github.com/tBeltty/auditor-executor-protocol/tree/main/integrations/claude-code/aeprogress).
+It works in the terminal and in the desktop app's local sessions (in the app: **+** next to the prompt, **Plugins**, **Add marketplace**). Every limit and feature is an option in `/config`. It needs Claude Code 2.1.287 or later and sends no data about you anywhere (its one network request, a daily version check, can be turned off); its README lists every process it starts and everything it can refuse. To get new versions on their own, turn on auto-update once: `/plugin`, **Marketplaces**, `auditor-executor-protocol`, **Enable auto-update**. Setup, options and how each number is computed: [integrations/claude-code/aeprogress](https://github.com/tBeltty/auditor-executor-protocol/tree/main/integrations/claude-code/aeprogress).
 
 ---
 

{aepkit-0.7.3 → aepkit-0.7.4}/references/autonomous-mode.md

@@ -104,6 +104,16 @@ Do not end a turn between a verdict and the next launch unless something on the
 critical-risk list is open. A run that stops to report an `APPROVED` and wait is Guided
 mode by accident.
 
+### Tool versions at the start of a run
+
+Before launching the first task, the Auditor updates the tools the run depends on (`aepkit`
+with `pipx upgrade aepkit`; any plugin or harness the run uses, such as the optional
+`aeprogress` mod with `claude plugin update aeprogress@auditor-executor-protocol`) and
+records the versions it ends up on as a governance decision in the plan of record. A run
+then uses one set of tools from its first task to its last, and whoever reads the log
+later knows which `aepkit lint` rules and which launch limits judged it. Updating mid-run
+is a decision like any other: record it, with the reason.
+
 ### Starting the Auditor (the Auditor brief)
 
 An autonomous run is started by giving a fresh Auditor session a brief, the same way an
@@ -132,6 +142,9 @@ DOCUMENTS: <plan of record, guide, log — what each contains and who writes whe
 
 STATE: <local/pushed commits, clean tree or not, first task and its model>.
 
+TOOLS: <the tools the run depends on and the versions to update to and record before task 1
+(aepkit; any plugin or harness), per "Tool versions at the start of a run">.
+
 GOAL (non-negotiable): <the owner's objective, in their own words>. Splitting into
 phases, yes; shrinking it, no. RUN COMPLETE only with <final gate> APPROVED.
 

aepkit-0.7.4/src/aepkit/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.7.4"

{aepkit-0.7.3 → aepkit-0.7.4}/src/aepkit/skill/references/autonomous-mode.md

@@ -104,6 +104,16 @@ Do not end a turn between a verdict and the next launch unless something on the
 critical-risk list is open. A run that stops to report an `APPROVED` and wait is Guided
 mode by accident.
 
+### Tool versions at the start of a run
+
+Before launching the first task, the Auditor updates the tools the run depends on (`aepkit`
+with `pipx upgrade aepkit`; any plugin or harness the run uses, such as the optional
+`aeprogress` mod with `claude plugin update aeprogress@auditor-executor-protocol`) and
+records the versions it ends up on as a governance decision in the plan of record. A run
+then uses one set of tools from its first task to its last, and whoever reads the log
+later knows which `aepkit lint` rules and which launch limits judged it. Updating mid-run
+is a decision like any other: record it, with the reason.
+
 ### Starting the Auditor (the Auditor brief)
 
 An autonomous run is started by giving a fresh Auditor session a brief, the same way an
@@ -132,6 +142,9 @@ DOCUMENTS: <plan of record, guide, log — what each contains and who writes whe
 
 STATE: <local/pushed commits, clean tree or not, first task and its model>.
 
+TOOLS: <the tools the run depends on and the versions to update to and record before task 1
+(aepkit; any plugin or harness), per "Tool versions at the start of a run">.
+
 GOAL (non-negotiable): <the owner's objective, in their own words>. Splitting into
 phases, yes; shrinking it, no. RUN COMPLETE only with <final gate> APPROVED.
 

{aepkit-0.7.3 → aepkit-0.7.4/src/aepkit.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: aepkit
-Version: 0.7.3
+Version: 0.7.4
 Summary: Mechanical support for the Auditor/Executor protocol: scaffold, lint, and negative-control the document set for a multi-phase, multi-agent execution.
 Author-email: tBeltty <jhonatan@tbelt.online>
 License-Expression: MIT
@@ -183,7 +183,7 @@ A failure only proves something if it is the failure you meant. A syntax error,
 
 `aeprogress` is a Claude Code mod for runs where the Auditor launches Executors as subagents or as workers of a harness such as Orca. The protocol does not depend on it, and it reads the run only through `aepkit status --json`.
 
-<img src="https://raw.githubusercontent.com/tBeltty/auditor-executor-protocol/main/integrations/claude-code/aeprogress/assets/pane.png" alt="The aeprogress pane: 43.3% overall, a bar per phase, P2-T3 in progress on a subagent, P2-T4 awaiting audit, P2-T2 conditional, and the machine at 31% with 1 of 2 workers live" width="420">
+<img src="https://raw.githubusercontent.com/tBeltty/auditor-executor-protocol/main/integrations/claude-code/aeprogress/assets/pane.png" alt="The aeprogress pane: 43.3% overall, a bar per phase, the active phase P2 open with its task chips, P2-T3 in progress on a subagent, P2-T4 awaiting audit and P2-T2 conditional" width="420">
 
 It does three things:
 
@@ -203,7 +203,7 @@ Install it at the prompt of a Claude Code terminal session, then open it:
 /aeprogress docs/<task-name>
 ```
 
-It works in the terminal and in the desktop app's local sessions (in the app: **+** next to the prompt, **Plugins**, **Add marketplace**). Every limit and feature is an option in `/config`. Claude Code marks its mod API as early access, so a Claude Code release can break it. Setup, options and how each number is computed: [integrations/claude-code/aeprogress](https://github.com/tBeltty/auditor-executor-protocol/tree/main/integrations/claude-code/aeprogress).
+It works in the terminal and in the desktop app's local sessions (in the app: **+** next to the prompt, **Plugins**, **Add marketplace**). Every limit and feature is an option in `/config`. It needs Claude Code 2.1.287 or later and sends no data about you anywhere (its one network request, a daily version check, can be turned off); its README lists every process it starts and everything it can refuse. To get new versions on their own, turn on auto-update once: `/plugin`, **Marketplaces**, `auditor-executor-protocol`, **Enable auto-update**. Setup, options and how each number is computed: [integrations/claude-code/aeprogress](https://github.com/tBeltty/auditor-executor-protocol/tree/main/integrations/claude-code/aeprogress).
 
 ---
 

aepkit-0.7.3/src/aepkit/__init__.py

@@ -1 +0,0 @@
-__version__ = "0.7.3"