Connect wrapper MVP outputs to governor docs
This commit is contained in:
@@ -41,7 +41,7 @@ Must explicitly say what the assistant will report if the task is not done yet.
|
||||
|
||||
### `externalized_trigger`
|
||||
Examples:
|
||||
- cron reminder in 20 minutes
|
||||
- cron reminder
|
||||
- forced checkpoint scheduled externally
|
||||
- non-silent mode (if no external trigger can safely be bound, the task should not actually launch as silent)
|
||||
|
||||
@@ -50,6 +50,18 @@ Examples:
|
||||
- Telegram button-path
|
||||
- direct conclusion if no owner decision needed
|
||||
|
||||
## Wrapper mapping
|
||||
|
||||
If using `scripts/long_task_governor_wrapper.mjs`, map wrapper output into this template like this:
|
||||
|
||||
- `classification = long_task` → task should use this launch template
|
||||
- `silentCandidate = true` → fill `why_silent`, `first_forced_checkpoint_trigger`, `externalized_trigger`
|
||||
- `taskRecord.current_step` → `current_step`
|
||||
- `taskRecord.next_step` → `next_step`
|
||||
- `taskRecord.status` → `current_status`
|
||||
- `silentLaunchOk = false` → do **not** launch in silent mode; fall back to immediate follow-up or `blocked`
|
||||
- `handoff.mode = button_path` → `final_handoff_path` should be Telegram button-path
|
||||
|
||||
## Telegram-safe example
|
||||
|
||||
- **task_name**: Research OpenClawGovernor concepts
|
||||
|
||||
@@ -110,6 +110,15 @@ When entering long-task governance, create or update a record with at least:
|
||||
|
||||
Use the template file `task-record-template.md` in this skill directory.
|
||||
|
||||
If using the wrapper MVP (`scripts/long_task_governor_wrapper.mjs`), consume:
|
||||
- `classification`
|
||||
- `silentCandidate`
|
||||
- `taskRecord`
|
||||
- `silentLaunchOk`
|
||||
- `handoff.mode`
|
||||
|
||||
as the first machine-readable pass before applying the human-facing workflow.
|
||||
|
||||
---
|
||||
|
||||
## 4. Checkpoint protocol
|
||||
@@ -154,6 +163,15 @@ If no externalized checkpoint path can be created safely, do not launch the task
|
||||
For reminder payload design, use:
|
||||
- `docs/runbooks/silent-long-task-reminder-contract.md`
|
||||
|
||||
For launch decisions, use:
|
||||
- `docs/runbooks/silent-long-task-decision-tree.md`
|
||||
- `docs/runbooks/silent-long-task-launch-template.md`
|
||||
|
||||
If using the wrapper MVP:
|
||||
- `silentCandidate = true` means you must evaluate silent launch rules
|
||||
- `silentLaunchOk = false` means you must not proceed as silent
|
||||
- `handoff.mode = button_path` means prepare Telegram button-path early
|
||||
|
||||
---
|
||||
|
||||
## 6. No-fake-progress rule
|
||||
@@ -193,8 +211,8 @@ Without that, default to `paused`.
|
||||
|
||||
When this skill applies:
|
||||
1. decide if the request is ordinary chat or long task
|
||||
2. if long task, create/update a task record
|
||||
3. choose one of the five valid states
|
||||
2. if available, run the wrapper MVP first as the machine-readable classification/bootstrap layer
|
||||
3. if long task, create/update a task record
|
||||
4. if the task is silent, define the first forced checkpoint before proceeding
|
||||
5. if the task is silent, externalize the checkpoint path or keep the task non-silent
|
||||
6. if reporting progress, use the 5-part checkpoint structure
|
||||
|
||||
Reference in New Issue
Block a user