cowork-flow 0.0.12 → 0.0.14
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/package.json
CHANGED
|
@@ -39,8 +39,10 @@ Do not invoke `superpowers:subagent-driven-development` from this skill. Agent-t
|
|
|
39
39
|
### Fresh Worker Per Assignment
|
|
40
40
|
|
|
41
41
|
- 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.
|
|
42
|
+
- Do not request a full-history fork when you also specify `agent_type`. In Codex hosts that expose real subagent dispatch tools, full-history fork cannot be combined with `agent_type` and the spawn call will be rejected.
|
|
42
43
|
- 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.
|
|
43
44
|
- Spawn a fresh worker for each ready assignment. Do not reuse the main agent context as the worker context.
|
|
45
|
+
- When the host offers fork controls, prefer a fresh child thread with explicit prompt context over inherited full history. Put the assignment Markdown, write boundary, and any needed task facts directly in the worker prompt.
|
|
44
46
|
- Use `agent-team/assignments/<assignment-id>.md` as the worker prompt body.
|
|
45
47
|
- 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.
|
|
46
48
|
- 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.
|
|
@@ -3,6 +3,10 @@ name: start
|
|
|
3
3
|
description: Use when starting or resuming work in a project that uses the cowork-flow template, after context compression, or before any request that may modify repository files. Initialize session context and route work into the L0/L1/L2 workflow.
|
|
4
4
|
---
|
|
5
5
|
|
|
6
|
+
<SUBAGENT-STOP>
|
|
7
|
+
If you were dispatched as a subagent to execute a specific assignment, skip this skill.
|
|
8
|
+
</SUBAGENT-STOP>
|
|
9
|
+
|
|
6
10
|
# Start Session
|
|
7
11
|
|
|
8
12
|
Use this skill as the entrypoint for cowork-flow projects.
|
|
@@ -80,4 +84,3 @@ When resuming after a long task, a new session, or context compression:
|
|
|
80
84
|
2. Follow `RESUME CHECKLIST` from `resume.py`.
|
|
81
85
|
3. Read only the current PRD, current plan status, and the jsonl references needed for the current phase.
|
|
82
86
|
4. Do not bulk-read `.cowork-flow/spec/`, all plans, all tasks, or workspace journals.
|
|
83
|
-
|
|
@@ -1 +1 @@
|
|
|
1
|
-
0.0.
|
|
1
|
+
0.0.14
|
|
@@ -479,6 +479,10 @@ def render_assignment_prompt(assignment: dict[str, object]) -> str:
|
|
|
479
479
|
"",
|
|
480
480
|
f"Spawn one {agent_type} agent for this assignment. Use this file as the complete worker prompt.",
|
|
481
481
|
"",
|
|
482
|
+
"You are already the dispatched worker for this assignment.",
|
|
483
|
+
"Do not run the project start-session workflow or try to spawn another worker from this prompt.",
|
|
484
|
+
"Treat this file as the complete worker brief unless you are blocked and need one specific missing fact.",
|
|
485
|
+
"",
|
|
482
486
|
]
|
|
483
487
|
agent_prompt = assignment.get("agent_prompt")
|
|
484
488
|
if isinstance(agent_prompt, str) and agent_prompt.strip():
|