cowork-flow 0.0.8 → 0.0.11

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "cowork-flow",
-  "version": "0.0.8",
+  "version": "0.0.11",
   "description": "CLI for installing and maintaining cowork-flow collaboration templates.",
   "type": "module",
   "bin": {

package/template/.agent/skills/agent-team-execution/SKILL.md

@@ -10,17 +10,50 @@ Use this skill after a task is active and an approved `.cowork-flow/plans/*.md`
 ## Process
 
 1. Run `./.cowork-flow/run agent-team prepare <task-dir> --plan <plan-file>`.
-2. Review `agent-team/dispatch-plan.yaml` for unsafe parallelism, file conflicts, missing context, or weak agent matches.
+2. Review `agent-team/dispatch-plan.yaml` for unsafe parallelism, file conflicts, missing context, or weak agent matches. If the graph is unsafe, fix the plan dependencies or split assignments, then rerun `prepare`.
 3. Run `./.cowork-flow/run agent-team next <task-dir>` to get ready assignments.
-4. In Codex, dispatch the ready assignments to the recommended agent types. Tell each worker it is not alone in the codebase, must respect the write boundary, must not revert others' edits, and must list changed files.
+4. Dispatch ready assignments using the Dispatch Protocol below.
 5. While workers run, coordinate: answer questions, unblock context gaps, and integrate non-conflicting results.
-6. Record outputs with `record-result` and reviews with `record-review`.
+6. Record implementer outputs with `record-result` and reviewer outputs with `record-review`.
 7. Use `retry` only after adding missing context, changing agent choice, or splitting an oversized assignment.
-8. Run `complete` before claiming the agent team work is done.
+8. Repeat `next -> dispatch -> record` until no assignments remain.
+9. Run `./.cowork-flow/run agent-team complete <task-dir>` before claiming the agent team work is done.
+
+### Codex Orchestration Prompt
+
+- Ask Codex explicitly to spawn subagents. Use this wording for each ready batch: `Spawn one <agent_type> agent per ready assignment, wait for all requested results, and summarize each result by assignment id.`
+- Codex handles spawning, routing follow-up instructions, waiting for results, and closing completed agent threads. Do not look for an `agent-team` spawn subcommand; the Python script only prepares state and prompts.
+- Use `/agent` in interactive CLI sessions when you need to inspect, steer, stop, or close active agent threads.
+
+### Fresh Worker Per Assignment
+
+- Spawn one agent per ready assignment. Use the assignment's `agent_type` as the Codex spawn target; built-in Codex agent types include `default`, `worker`, and `explorer`, and project or personal custom agents may add more.
+- Treat `recommended_agent` as the agent-team registry match and prompt source, not as the Codex spawn target unless it also names a real Codex custom agent.
+- Spawn a fresh worker for each ready assignment. Do not reuse the main agent context as the worker context.
+- Use `agent-team/assignments/<assignment-id>.md` as the worker prompt body.
+- Add this scene-setting to every worker prompt: it is not alone in the codebase, it must respect the listed write boundary, it must not revert other agents' edits, and it must report changed files plus exact verification commands.
+- Dispatch multiple ready assignments in parallel only after confirming their write files do not overlap. If they overlap, dispatch them sequentially or repair the dependency graph.
+
+### Worker Result Handling
+
+- Wait for the consolidated Codex response for the requested batch before recording results.
+- Treat worker statuses as follows:
+  - `DONE`: inspect changed files and verification evidence, then record `done` with `record-result`.
+  - `DONE_WITH_CONCERNS`: inspect the concerns before recording; retry if they affect correctness or scope.
+  - `NEEDS_CONTEXT`: add the missing context to the assignment or task context, then use `retry`.
+  - `BLOCKED`: decide whether to add context, choose a stronger agent, split the assignment, or stop for coordinator decision.
+
+### Review Chain
+
+- After an implementer is recorded, let `agent-team next` unlock that task's spec reviewer. Dispatch the reviewer as a fresh worker using its generated assignment prompt.
+- If the spec reviewer rejects or flags gaps, record the review result, retry the implementer with the specific gaps, then rerun the review.
+- Only after spec review is approved should the quality reviewer run. If quality review rejects, retry the implementer with the specific issues and rerun the review chain.
+- Do not let implementer self-review replace `record-review` for spec or quality assignments.
 
 ## Rules
 
 - The script suggests; the main agent decides.
 - Do not parallelize assignments with overlapping write files.
 - Do not skip spec review or quality review.
+- In Codex, `agent-team` is coordinator-dispatched: the Python script generates assignments, and the main agent performs the real subagent dispatch calls while preserving `agent-team` state.
 - Do not rely on chat history for state; write results through `agent-team` commands.

package/template/.cowork-flow/.version

@@ -1 +1 @@
-0.0.8
+0.0.11

package/template/.cowork-flow/agent-team/agents.yaml

@@ -13,10 +13,6 @@ agents:
       - code
       - test
     file_patterns:
-      - "src/**"
-      - "template/**"
-      - "tests/**"
-      - "test/**"
     risk_limits:
       max_parallel_write_conflicts: 0
     prompt: |
@@ -32,10 +28,6 @@ agents:
       - test
       - verification
     file_patterns:
-      - "tests/**"
-      - "test/**"
-      - "src/**"
-      - "template/**"
     risk_limits:
       max_parallel_write_conflicts: 0
     prompt: |
@@ -52,10 +44,6 @@ agents:
       - bugfix
       - code
     file_patterns:
-      - "src/**"
-      - "template/**"
-      - "tests/**"
-      - "test/**"
     risk_limits:
       max_parallel_write_conflicts: 0
     prompt: |
@@ -71,10 +59,6 @@ agents:
       - review
       - documentation
     file_patterns:
-      - ".cowork-flow/changes/**"
-      - ".cowork-flow/plans/**"
-      - "README.md"
-      - "template/**"
     risk_limits:
       max_parallel_write_conflicts: 0
     prompt: |
@@ -91,10 +75,6 @@ agents:
       - review
       - verification
     file_patterns:
-      - "src/**"
-      - "template/**"
-      - "tests/**"
-      - "test/**"
     risk_limits:
       max_parallel_write_conflicts: 0
     prompt: |
@@ -110,10 +90,6 @@ agents:
       - docs
       - documentation
     file_patterns:
-      - "README.md"
-      - "AGENTS.md"
-      - ".cowork-flow/**/*.md"
-      - ".agent/skills/**"
     risk_limits:
       max_parallel_write_conflicts: 0
     prompt: |
@@ -130,11 +106,6 @@ agents:
       - release
       - review
     file_patterns:
-      - "package.json"
-      - "package-lock.json"
-      - "scripts/**"
-      - "README.md"
-      - ".cowork-flow/changes/**"
     risk_limits:
       max_parallel_write_conflicts: 0
     prompt: |

package/template/.cowork-flow/scripts/common/agent_team.py

@@ -7,9 +7,10 @@ import re
 from pathlib import Path
 
 
-TASK_RE = re.compile(r"^### Task\s+(\d+):\s*(.+?)\s*$")
+TASK_RE = re.compile(r"^#{2,3} Task\s+(\d+):\s*(.+?)\s*$")
 FILE_RE = re.compile(r"^-\s+(Create|Modify|Test):\s+`([^`]+)`")
 DEP_RE = re.compile(r"depends on Task\s+(\d+)", re.IGNORECASE)
+TERMINAL_TASK_RE = re.compile(r"(final|integration|verify|verification|验收|集成|最终|全量|收尾)", re.IGNORECASE)
 
 
 def parse_plan(text: str) -> list[dict[str, object]]:
@@ -99,6 +100,25 @@ def _task_dependencies(tasks: list[dict[str, object]]) -> list[dict[str, str]]:
                     }
                 )
 
+        title = task.get("title", "")
+        if isinstance(title, str) and TERMINAL_TASK_RE.search(title):
+            existing = {
+                dependency["depends_on_task"]
+                for dependency in dependencies
+                if dependency["task"] == task_id
+            }
+            for previous in tasks[:index]:
+                previous_id = str(previous["id"])
+                if previous_id in existing:
+                    continue
+                dependencies.append(
+                    {
+                        "task": task_id,
+                        "depends_on_task": previous_id,
+                        "reason": "terminal-task",
+                    }
+                )
+
     return dependencies
 
 
@@ -447,14 +467,18 @@ def _render_named_items(title: str, items: object) -> list[str]:
 
 
 def render_assignment_prompt(assignment: dict[str, object]) -> str:
+    agent_type = assignment["agent_type"]
     lines = [
         f"# {assignment['id']}",
         "",
         f"Role: {assignment['role']}",
         f"Recommended agent: {assignment['recommended_agent']}",
-        f"Agent type: {assignment['agent_type']}",
+        f"Agent type: {agent_type}",
+        f"Spawn target agent type: {agent_type}",
         f"Task: {assignment.get('title', '')}",
         "",
+        f"Spawn one {agent_type} agent for this assignment. Use this file as the complete worker prompt.",
+        "",
     ]
     agent_prompt = assignment.get("agent_prompt")
     if isinstance(agent_prompt, str) and agent_prompt.strip():