ouroboros-ai 0.30.1.dev4__tar.gz → 0.30.1.dev7__tar.gz

{ouroboros_ai-0.30.1.dev4 → ouroboros_ai-0.30.1.dev7}/CHANGELOG.md

@@ -13,6 +13,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - **opencode/bridge**: Plugin v23 recognizes `_subagents` array for parallel fan-out. Per-payload validation, truncation, and dedupe. One failed dispatch does not abort the rest. New `ouroboros_subagents` and `ouroboros_dispatch_errors` metadata fields. Backwards compatible with v22 single-payload `_subagent` contract.
 
 ### Fixed
+- **skills**: Renamed the packaged `resume` skill to `resume-session` so Claude Code's built-in `/resume` session picker is no longer shadowed. Use `ooo resume-session` or `/ouroboros:resume-session` for the Ouroboros in-flight session listing.
 - **mcp/security**: `FREETEXT_FIELDS` allowlist for user-input fields (goals, prompts, descriptions) — shell metacharacters (`;`, `|`, `&`, backticks, `$()`) are no longer rejected in fields where they are legitimate prose. Structural fields remain strictly validated.
 - **opencode/bridge**: Robustness hardening (v22) — no uncaught errors under any input. Adds reject-path logging, frozen-content guards, empty-sessionID guard, client init-order guard, 5-second FNV-1a prompt dedupe, 100 KB prompt byte cap with truncation marker, user-visible `surfaceErr()` for dispatch failures (no more silent "dispatched but never ran"), and an absolute outer try/catch so the plugin cannot throw into the opencode runLoop.
 

{ouroboros_ai-0.30.1.dev4 → ouroboros_ai-0.30.1.dev7}/CLAUDE.md

@@ -32,7 +32,7 @@ When the user types any of these commands, read the corresponding SKILL.md file
 | `ooo pm` or `ooo pm ...` | Read `skills/pm/SKILL.md` and follow it |
 | `ooo brownfield` or `ooo brownfield ...` | Read `skills/brownfield/SKILL.md` and follow it |
 | `ooo publish` or `ooo publish ...` | Read `skills/publish/SKILL.md` and follow it |
-| `ooo resume` | Read `skills/resume/SKILL.md` and follow it |
+| `ooo resume-session` | Read `skills/resume-session/SKILL.md` and follow it |
 
 **Important**: Do NOT use the Skill tool. Read the file with the Read tool and execute its instructions directly.
 
@@ -82,6 +82,7 @@ Each command loads its agent/MCP on-demand. Details in each skill file.
 | `ooo help` | — |
 | `ooo update` | PyPI version check + upgrade |
 | `ooo publish` | `gh` CLI — Seed to GitHub Issues |
+| `ooo resume-session` | Restore previous Claude Code session context |
 
 ## Agents
 

{ouroboros_ai-0.30.1.dev4 → ouroboros_ai-0.30.1.dev7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ouroboros-ai
-Version: 0.30.1.dev4
+Version: 0.30.1.dev7
 Summary: Specification-first workflow engine for AI coding agents. Works with Claude Code and Codex CLI.
 Author-email: Q00 <jqyu.lee@gmail.com>
 License-File: LICENSE
@@ -250,6 +250,7 @@ Inside AI coding agent sessions, use `ooo <cmd>` skills. From the terminal, use
 | `ooo evolve`     | *(via MCP)*                                                       | Evolutionary loop until ontology converges                   |
 | `ooo unstuck`    | *(via MCP)*                                                       | 5 lateral thinking personas when you are stuck               |
 | `ooo status`     | `ouroboros status executions` / `ouroboros status execution <id>` | Session tracking + (MCP-only) drift detection                |
+| `ooo resume-session` | `ouroboros resume`                                           | List in-flight sessions and re-attach commands              |
 | `ooo cancel`     | `ouroboros cancel execution [<id>\|--all]`                        | Cancel stuck or orphaned executions                          |
 | `ooo ralph`      | *(via MCP)*                                                       | Persistent loop until verified                               |
 | `ooo tutorial`   | *(interactive)*                                                   | Interactive hands-on learning                                |
@@ -261,6 +262,7 @@ Inside AI coding agent sessions, use `ooo <cmd>` skills. From the terminal, use
 | `ooo publish`    | *(skill/runtime surface; uses `gh` CLI)*                          | Publish a Seed as GitHub Epic/Task issues for team workflows |
 
 > Not all skills have direct CLI equivalents. Some (`evaluate`, `evolve`, `unstuck`, `ralph`, `publish`) are available through agent skills, runtime rules, or MCP tools rather than a direct `ouroboros <subcommand>` shell command.
+> `/resume` is reserved for Claude Code's built-in session picker; use `ooo resume-session` for Ouroboros in-flight sessions.
 
 See the [CLI reference](./docs/cli-reference.md) for full details.
 

{ouroboros_ai-0.30.1.dev4 → ouroboros_ai-0.30.1.dev7}/README.ko.md

@@ -267,6 +267,7 @@ Gen 3: {Task, Priority, Status, DueDate}     → similarity 1.00 → CONVERGED
 | `ooo evolve` | 온톨로지 수렴까지 진화 루프 |
 | `ooo unstuck` | 막혔을 때 활용 가능한 5가지 수평적 사고 페르소나 |
 | `ooo status` | 드리프트 감지 + 세션 추적 |
+| `ooo resume-session` | 실행 중인 세션 목록과 재연결 명령 확인 |
 | `ooo ralph` | 검증될 때까지 계속 도는 루프 |
 | `ooo tutorial` | 대화형 실습 |
 | `ooo help` | 전체 참조 |

{ouroboros_ai-0.30.1.dev4 → ouroboros_ai-0.30.1.dev7}/README.md

@@ -210,6 +210,7 @@ Inside AI coding agent sessions, use `ooo <cmd>` skills. From the terminal, use
 | `ooo evolve`     | *(via MCP)*                                                       | Evolutionary loop until ontology converges                   |
 | `ooo unstuck`    | *(via MCP)*                                                       | 5 lateral thinking personas when you are stuck               |
 | `ooo status`     | `ouroboros status executions` / `ouroboros status execution <id>` | Session tracking + (MCP-only) drift detection                |
+| `ooo resume-session` | `ouroboros resume`                                           | List in-flight sessions and re-attach commands              |
 | `ooo cancel`     | `ouroboros cancel execution [<id>\|--all]`                        | Cancel stuck or orphaned executions                          |
 | `ooo ralph`      | *(via MCP)*                                                       | Persistent loop until verified                               |
 | `ooo tutorial`   | *(interactive)*                                                   | Interactive hands-on learning                                |
@@ -221,6 +222,7 @@ Inside AI coding agent sessions, use `ooo <cmd>` skills. From the terminal, use
 | `ooo publish`    | *(skill/runtime surface; uses `gh` CLI)*                          | Publish a Seed as GitHub Epic/Task issues for team workflows |
 
 > Not all skills have direct CLI equivalents. Some (`evaluate`, `evolve`, `unstuck`, `ralph`, `publish`) are available through agent skills, runtime rules, or MCP tools rather than a direct `ouroboros <subcommand>` shell command.
+> `/resume` is reserved for Claude Code's built-in session picker; use `ooo resume-session` for Ouroboros in-flight sessions.
 
 See the [CLI reference](./docs/cli-reference.md) for full details.
 

{ouroboros_ai-0.30.1.dev4 → ouroboros_ai-0.30.1.dev7}/docs/events.md

@@ -1,7 +1,7 @@
 # Event Payload Schema Reference
 
 This document defines the stable payload fields for Ouroboros EventStore
-events. Consumers that read events -- TUI, `ooo status`, `ooo resume`,
+events. Consumers that read events -- TUI, `ooo status`, `ooo resume-session`,
 `ouroboros_query_events` -- can rely on these
 fields not being removed or renamed within a given `event_version`.
 

{ouroboros_ai-0.30.1.dev4 → ouroboros_ai-0.30.1.dev7}/skills/help/SKILL.md

@@ -37,6 +37,7 @@ Ouroboros is a **requirement crystallization engine** for AI workflows. It trans
 | `ooo evaluate` | 3-stage verification | MCP |
 | `ooo unstuck` | 5 lateral thinking personas | Plugin |
 | `ooo status` | Session status + drift check | MCP |
+| `ooo resume-session` | List in-flight sessions and re-attach commands | CLI |
 | `ooo setup` | Installation wizard | Plugin |
 | `ooo welcome` | First-touch welcome guide | Plugin |
 | `ooo tutorial` | Interactive hands-on learning | Plugin |
@@ -76,6 +77,7 @@ Ouroboros is a **requirement crystallization engine** for AI workflows. It trans
 | "write prd", "pm interview", "product requirements", "create prd" | `ooo pm` |
 | "qa check", "quality check" | `ooo qa` |
 | "cancel execution", "stop job", "kill stuck", "abort execution" | `ooo cancel` |
+| "in-flight sessions", "mcp disconnected", "lost Ouroboros execution" | `ooo resume-session` |
 | "update ouroboros", "upgrade ouroboros" | `ooo update` |
 | "brownfield defaults", "brownfield scan" | `ooo brownfield` |
 | "publish to github", "create issues from seed", "seed to issues" | `ooo publish` |
@@ -100,6 +102,7 @@ Ouroboros is a **requirement crystallization engine** for AI workflows. It trans
 | `/ouroboros:evaluate` | 3-stage verification | MCP |
 | `/ouroboros:unstuck` | 5 lateral thinking personas | Plugin |
 | `/ouroboros:status` | Session status + drift check | MCP |
+| `/ouroboros:resume-session` | List in-flight sessions and re-attach commands | CLI |
 | `/ouroboros:setup` | Installation wizard | Plugin |
 | `/ouroboros:tutorial` | Interactive hands-on learning | Plugin |
 | `/ouroboros:help` | This guide | Plugin |

{ouroboros_ai-0.30.1.dev4/skills/resume → ouroboros_ai-0.30.1.dev7/skills/resume-session}/SKILL.md

@@ -1,31 +1,35 @@
 ---
-name: resume
-description: "List in-flight sessions and show the commands needed to re-attach after MCP disconnect"
+name: resume-session
+description: "List in-flight Ouroboros sessions and show the commands needed to re-attach after MCP disconnect"
 ---
 
-# /ouroboros:resume
+# /ouroboros:resume-session
 
-Recover in-flight sessions after an unexpected MCP server disconnect.
+Recover in-flight Ouroboros sessions after an unexpected MCP server disconnect.
+
+Claude Code reserves `/resume` for its built-in session picker. This skill
+intentionally uses `resume-session` so it does not shadow that native command.
 
 ## Usage
 
 ```
-ooo resume
-ooo resume --all
+ooo resume-session
+ooo resume-session --all
+/ouroboros:resume-session
 ```
 
-**Trigger keywords:** "resume session", "re-attach", "mcp disconnected", "lost session", "in-flight"
+**Trigger keywords:** "in-flight Ouroboros sessions", "re-attach", "mcp disconnected", "lost Ouroboros execution"
 
 ## How It Works
 
-`ooo resume` reads the EventStore directly (no MCP server required) and lists
+`ooo resume-session` reads the EventStore directly (no MCP server required) and lists
 every session that is still in a `running` or `paused` state. The command is
 strictly read-only — it never creates the data directory, never writes
 schema, and never appends events. Its job is to surface the identifiers you
 need to re-attach.
 
-- `ooo resume` shows the 20 most recent active sessions.
-- `ooo resume --all` shows every active session.
+- `ooo resume-session` shows the 20 most recent active sessions.
+- `ooo resume-session --all` shows every active session.
 
 ## Instructions
 
@@ -85,7 +89,7 @@ ouroboros tui monitor
 ## Example
 
 ```
-User: ooo resume
+User: ooo resume-session
 
 ┌─────────────────────── In-Flight Sessions ───────────────────────┐
 │  #  Session ID          Execution ID        Status    Started    │

{ouroboros_ai-0.30.1.dev4 → ouroboros_ai-0.30.1.dev7}/src/ouroboros/_version.py

@@ -18,7 +18,7 @@ version_tuple: tuple[int | str, ...]
 commit_id: str | None
 __commit_id__: str | None
 
-__version__ = version = '0.30.1.dev4'
-__version_tuple__ = version_tuple = (0, 30, 1, 'dev4')
+__version__ = version = '0.30.1.dev7'
+__version_tuple__ = version_tuple = (0, 30, 1, 'dev7')
 
 __commit_id__ = commit_id = None

{ouroboros_ai-0.30.1.dev4 → ouroboros_ai-0.30.1.dev7}/src/ouroboros/execution/decomposition.py

@@ -56,7 +56,7 @@ log = get_logger(__name__)
 # Decomposition constraints
 MIN_CHILDREN = 2
 MAX_CHILDREN = 5
-MAX_DEPTH = 5
+MAX_DEPTH = 2
 COMPRESSION_DEPTH = 3
 
 

{ouroboros_ai-0.30.1.dev4 → ouroboros_ai-0.30.1.dev7}/tests/unit/cli/test_doc_commands.py

@@ -14,7 +14,7 @@ no MCP spin-up, no seed file required).
 Scope:
 
 - Every command line emitted by ``resume._format_reattach_guidance``.
-- Every ``ouroboros ...`` directive documented in ``skills/resume/SKILL.md``.
+- Every ``ouroboros ...`` directive documented in ``skills/resume-session/SKILL.md``.
 - The bare ``ouroboros`` entrypoints listed in the CLAUDE.md ``ooo`` table
   that correspond to a real CLI subcommand (plugin-only skills such as
   ``ooo welcome`` are not CLI commands and are skipped).
@@ -115,7 +115,7 @@ class TestReattachGuidanceCommandsAreReal:
 
 
 # ---------------------------------------------------------------------------
-# 2) Every `ouroboros ...` directive in skills/resume/SKILL.md
+# 2) Every `ouroboros ...` directive in skills/resume-session/SKILL.md
 # ---------------------------------------------------------------------------
 
 

{ouroboros_ai-0.30.1.dev4 → ouroboros_ai-0.30.1.dev7}/tests/unit/execution/test_decomposition.py

@@ -401,27 +401,20 @@ class TestDecomposeAc:
         assert result.is_err
         assert result.error.error_type == "insufficient_children"
 
-    @pytest.mark.asyncio
-    async def test_context_compression_at_depth(self, mock_llm_adapter):
-        """decompose_ac() should compress context at depth >= 3."""
-        from ouroboros.execution.decomposition import decompose_ac
+    def test_context_compression_at_depth(self):
+        """_compress_context should compress at depth >= COMPRESSION_DEPTH."""
+        from ouroboros.execution.decomposition import (
+            COMPRESSION_DEPTH,
+            _compress_context,
+        )
 
         long_insights = "A" * 1000
 
-        await decompose_ac(
-            ac_content="Test AC",
-            ac_id="ac_parent",
-            execution_id="exec_123",
-            depth=3,
-            llm_adapter=mock_llm_adapter,
-            discover_insights=long_insights,
-        )
+        compressed = _compress_context(long_insights, depth=COMPRESSION_DEPTH)
+        assert "[compressed for depth]" in compressed
 
-        # Check that LLM was called with compressed insights
-        call_args = mock_llm_adapter.complete.call_args
-        messages = call_args[0][0]
-        user_message = messages[1].content
-        assert "[compressed for depth]" in user_message
+        passthrough = _compress_context(long_insights, depth=COMPRESSION_DEPTH - 1)
+        assert passthrough == long_insights
 
     @pytest.mark.asyncio
     async def test_child_ids_are_unique(self, mock_llm_adapter):
@@ -458,10 +451,10 @@ class TestDecompositionConstants:
         assert MAX_CHILDREN == 5
 
     def test_max_depth_is_5(self):
-        """MAX_DEPTH should be 5."""
+        """MAX_DEPTH should be 2."""
         from ouroboros.execution.decomposition import MAX_DEPTH
 
-        assert MAX_DEPTH == 5
+        assert MAX_DEPTH == 2
 
     def test_compression_depth_is_3(self):
         """COMPRESSION_DEPTH should be 3."""

{ouroboros_ai-0.30.1.dev4 → ouroboros_ai-0.30.1.dev7}/tests/unit/router/test_registry.py

@@ -17,6 +17,7 @@ from ouroboros.router import (
     SkillDispatchTarget,
     SkillDispatchTargetResolution,
     normalize_skill_identifier,
+    packaged_skill_dispatch_registry,
     resolve_skill_dispatch,
     resolve_skill_dispatch_target,
 )
@@ -757,3 +758,14 @@ def test_packaged_registry_target_resolver_missing_skill_returns_not_handled(
 
 def test_packaged_registry_target_resolver_type_alias_includes_not_handled() -> None:
     assert SkillDispatchTargetResolution.__value__ == (SkillDispatchTarget | NotHandled)
+
+
+def test_packaged_registry_does_not_claim_claude_builtin_resume_command() -> None:
+    with packaged_skill_dispatch_registry() as registry:
+        reserved = registry.resolve("resume")
+        renamed = registry.resolve("resume-session")
+
+    assert isinstance(reserved, NotHandled)
+    assert reserved.category is NoMatchReason.SKILL_NOT_FOUND
+    assert isinstance(renamed, SkillDispatchTarget)
+    assert renamed.skill_name == "resume-session"