@pulsarinteractive/geniustasker 0.1.0-beta.12 → 0.1.0-beta.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/CHANGELOG.md +20 -0
- package/README.md +12 -1
- package/docs/CHATGPT.md +6 -0
- package/docs/CLI.md +75 -5
- package/docs/FEEDBACK.md +96 -0
- package/docs/INTEGRATIONS.md +25 -2
- package/docs/PLANNING.md +95 -0
- package/docs/TASK_LOCKS.md +91 -0
- package/examples/codex.toml +1 -1
- package/examples/mcp.json +1 -1
- package/mcp.json +1 -1
- package/package.json +1 -1
- package/plugin.json +1 -1
- package/remote/tool-catalog.json +779 -1
- package/remote/workflows.json +56 -2
- package/skills/geniustasker/SKILL.md +29 -1
- package/skills/geniustasker/references/agent-runs.md +13 -0
- package/skills/geniustasker/references/incremental-reads.md +9 -0
- package/skills/geniustasker/references/metric-goals.md +143 -0
- package/skills/geniustasker/references/metrics-observations.md +184 -0
- package/skills/geniustasker/references/metrics.md +136 -0
- package/skills/geniustasker/references/protocols.md +18 -3
- package/skills/geniustasker-feedback/SKILL.md +44 -0
- package/skills/geniustasker-feedback/agents/openai.yaml +4 -0
- package/skills/geniustasker-memory/SKILL.md +4 -1
- package/skills/geniustasker-planning/SKILL.md +28 -0
- package/src/changes.mjs +7 -5
- package/src/cli.mjs +165 -16
- package/src/client.mjs +16 -0
- package/src/command-catalog.mjs +151 -1
- package/src/command-input-schemas.mjs +1254 -93
- package/src/commands.mjs +257 -22
- package/src/feedback-contract.mjs +629 -0
- package/src/feedback-image.mjs +123 -0
- package/src/feedback-media-transfer.mjs +61 -0
- package/src/feedback.mjs +152 -0
- package/src/intents.mjs +111 -0
- package/src/mcp.mjs +143 -20
- package/src/protocols.mjs +1 -0
- package/src/resources.mjs +114 -0
- package/src/task-availability-contract.mjs +121 -0
- package/src/task-availability.mjs +25 -0
- package/src/transport.mjs +10 -3
package/CHANGELOG.md
CHANGED
|
@@ -1,5 +1,25 @@
|
|
|
1
1
|
# Changelog
|
|
2
2
|
|
|
3
|
+
## 0.1.0-beta.14 — 2026-11-17
|
|
4
|
+
|
|
5
|
+
- Add generated task create/update help and task/run lock commands without expanding task permissions. Guides distinguish canonical progress from current availability, preserve dependency locks after manual unlock and require the live run path for a claimed task. Availability and deployment remain server-authoritative.
|
|
6
|
+
- Read task availability only through explicit `taskState: "v1"` / `--task-state v1` after capability discovery. Legacy reads retain their old response shape. Validate optional task snapshots without treating an absent or incomplete source as available.
|
|
7
|
+
|
|
8
|
+
- Preserve optional protocol effort in command help and preview, including `0.1` working day and null for unknown. Teach truthful New/InProgress/Done progress, effort distinct from deadlines or AI runtime, and preservation of human estimates. Older deployments remain authoritative.
|
|
9
|
+
|
|
10
|
+
- Add private feedback to CLI, local MCP and the hosted tool catalogue: immutable bug/suggestion submission, receipts and final decision tracking without chat. Current grants, account plan and deployment activation remain authoritative.
|
|
11
|
+
- Add authorized local screenshot preparation, reservation, bounded binary transfer and exact retry. Hosted ChatGPT screenshot import is not implemented. A focused feedback skill requires the user's reporting instruction or agreement before submission.
|
|
12
|
+
|
|
13
|
+
- Leave default browser/device login scopes to the selected server's consent proposal so a newer CLI can connect to an earlier deployment. Explicit `--scope` remains unchanged; existing connections and profiles are never expanded automatically.
|
|
14
|
+
- Add bounded personal focus/defer reads and replacement previews to CLI, local MCP and the hosted tool catalogue. Explicit `intent.change` commands reuse current revisions and the immutable command journal; they do not change shared task dates, status or rewards.
|
|
15
|
+
- Document whole-selection replacement, cancellation, target invalidation and the connected owner boundary. Use only when `intent.change` is advertised by the deployment capabilities.
|
|
16
|
+
|
|
17
|
+
## 0.1.0-beta.13 — 2026-11-16
|
|
18
|
+
|
|
19
|
+
- Add personal and project custom metrics with scoped discovery, exact aggregation, observation history, corrections and voids.
|
|
20
|
+
- Link metrics to lifetime or fixed-window goals with bounded backfill, exact target comparisons and explicit notification/coach choices.
|
|
21
|
+
- Provide metric and goal guides, generated command schemas and a dedicated MCP write tool. Read current service capabilities and permissions before use; availability depends on the connected deployment.
|
|
22
|
+
|
|
3
23
|
## 0.1.0-beta.12 — 2026-11-16
|
|
4
24
|
|
|
5
25
|
- Add an opt-in host event adapter to CLI/local MCP with stable IDs/times, prepare-only support and the normal immutable journal. Process exit requests review, never task completion; no process/log capture or automatic polling.
|
package/README.md
CHANGED
|
@@ -19,6 +19,13 @@ Sign in in the browser, select your projects or memories, then approve the reque
|
|
|
19
19
|
permissions. The terminal connects automatically. No token to copy.
|
|
20
20
|
Installing the package does not grant application access.
|
|
21
21
|
|
|
22
|
+
The beta CLI and bundled local MCP examples default to **development**.
|
|
23
|
+
Production access is enabled for admitted accounts and is the environment used
|
|
24
|
+
by TestFlight. To work with that data, use `--environment production` and a
|
|
25
|
+
separate context at login and on subsequent commands and MCP configuration.
|
|
26
|
+
See [environment selection](docs/CLI.md#install-and-connect); a context name
|
|
27
|
+
alone does not change the environment.
|
|
28
|
+
|
|
22
29
|
## What you can do
|
|
23
30
|
|
|
24
31
|
| Area | Features |
|
|
@@ -48,7 +55,7 @@ For a stdio MCP host, use this after signing in:
|
|
|
48
55
|
"mcpServers": {
|
|
49
56
|
"geniustasker": {
|
|
50
57
|
"command": "npx",
|
|
51
|
-
"args": ["-y", "@pulsarinteractive/geniustasker@0.1.0-beta.
|
|
58
|
+
"args": ["-y", "@pulsarinteractive/geniustasker@0.1.0-beta.14", "mcp", "--context", "work"]
|
|
52
59
|
}
|
|
53
60
|
}
|
|
54
61
|
}
|
|
@@ -81,3 +88,7 @@ cover included components. The GeniusTasker application and backend are separate
|
|
|
81
88
|
## ChatGPT
|
|
82
89
|
|
|
83
90
|
Connect the hosted service with OAuth: [setup and examples](docs/CHATGPT.md). Your computer does not need to run the CLI. Account permissions and quotas still apply. [Official directory publication](docs/OPENAI_PUBLICATION.md) is a separate review.
|
|
91
|
+
|
|
92
|
+
## Feedback
|
|
93
|
+
|
|
94
|
+
Use [the feedback guide](docs/FEEDBACK.md) for authorized bug reports, feature suggestions, screenshots and final decisions. Availability follows current server capabilities, grants and plan. Preparing a report does not send it; submit after the user’s reporting instruction or agreement.
|
package/docs/CHATGPT.md
CHANGED
|
@@ -2,6 +2,12 @@
|
|
|
2
2
|
|
|
3
3
|
Use the hosted MCP connector to work with GeniusTasker from ChatGPT. Your computer does not need to run the CLI. Access still requires an admitted GeniusTasker account.
|
|
4
4
|
|
|
5
|
+
The URL below targets production, where agent access is enabled for admitted
|
|
6
|
+
accounts. This is the same environment used by TestFlight. Sign in with the
|
|
7
|
+
intended account and review its resource permissions. The beta CLI's development
|
|
8
|
+
default does not change this hosted connector's environment, and connecting a
|
|
9
|
+
custom server does not imply approval in OpenAI's public directory.
|
|
10
|
+
|
|
5
11
|
## Connect your account
|
|
6
12
|
|
|
7
13
|
1. In ChatGPT on the web, open **Plugins → Add → Add custom MCP server**.
|
package/docs/CLI.md
CHANGED
|
@@ -5,6 +5,28 @@ comment, ask questions and keep a useful handoff. Your account and project
|
|
|
5
5
|
permissions remain the limit. GeniusTasker does not run your agent or bill its
|
|
6
6
|
external model tokens.
|
|
7
7
|
|
|
8
|
+
## Custom metric definitions
|
|
9
|
+
|
|
10
|
+
For custom measurement definitions, use `metrics list --scope project:ID --json`
|
|
11
|
+
or `metrics list --scope personal:OWNER --json` (`OWNER` is identity's
|
|
12
|
+
`actor.ownerId`). Current capabilities must advertise the feature and the
|
|
13
|
+
separate `metrics:read` / `metrics:write` grants. `metric.create`,
|
|
14
|
+
`metric.update`, `metric.delete`, `metric.record`, `metric.correct` and
|
|
15
|
+
`metric.void`, plus the discoverable `metric.goal.create/update/delete` commands,
|
|
16
|
+
use the standard `commands prepare/execute/retry` journal.
|
|
17
|
+
MCP exposes `tasker_read_metrics` and `tasker_edit_metrics`.
|
|
18
|
+
Read the [metric configuration guide](../skills/geniustasker/references/metrics.md)
|
|
19
|
+
or `workflow show --topic metrics --json` before choosing the measurement meaning.
|
|
20
|
+
For observations or previous correction snapshots, choose `--collection`
|
|
21
|
+
explicitly; definitions remain the default. Read the
|
|
22
|
+
[observation guide](../skills/geniustasker/references/metrics-observations.md)
|
|
23
|
+
before recording or correcting evidence. For linked goals, select `metricGoals`
|
|
24
|
+
or `personalMetricGoals` and read the [metric goal guide](../skills/geniustasker/references/metric-goals.md).
|
|
25
|
+
A goal is a separate explicit command; definitions and observations never create
|
|
26
|
+
one implicitly. Goal thresholds are decimal strings, fixed windows use source
|
|
27
|
+
observation time, and `Pending` progress is not final. None of these commands
|
|
28
|
+
starts an automatic collector or executes an external agent.
|
|
29
|
+
|
|
8
30
|
## Task protocols
|
|
9
31
|
|
|
10
32
|
Discover `taskTemplates`, `protocolInstances` and the `protocol.*` commands in
|
|
@@ -46,10 +68,28 @@ permissions, then approve the connection. The CLI receives a one-use code with
|
|
|
46
68
|
PKCE; it never asks for your password. The consent page does not load the app tutorial. Decline there or use Ctrl-C to stop.
|
|
47
69
|
If the browser cannot open, use the link printed in your terminal.
|
|
48
70
|
|
|
49
|
-
**This beta defaults to the development environment.**
|
|
50
|
-
|
|
51
|
-
|
|
52
|
-
|
|
71
|
+
**This beta defaults to the development environment.** Production agent access
|
|
72
|
+
is enabled for admitted accounts; installation alone does not grant access in
|
|
73
|
+
either environment. Account admission, project membership, subscription and
|
|
74
|
+
browser consent still apply. No subscription purchase is performed by the CLI.
|
|
75
|
+
|
|
76
|
+
TestFlight uses production. To work on that same application's data, choose a
|
|
77
|
+
separate context and pass `--environment production` at login **and on every
|
|
78
|
+
subsequent CLI or local MCP invocation**:
|
|
79
|
+
|
|
80
|
+
```sh
|
|
81
|
+
geniustasker auth login --environment production --context tasker-beta --browser chrome
|
|
82
|
+
geniustasker whoami --environment production --context tasker-beta --json
|
|
83
|
+
geniustasker projects list --environment production --context tasker-beta --json
|
|
84
|
+
geniustasker mcp --environment production --context tasker-beta
|
|
85
|
+
```
|
|
86
|
+
|
|
87
|
+
A context name does not select an environment. Existing development sessions do
|
|
88
|
+
not become production sessions, and changing an environment does not migrate
|
|
89
|
+
records or expand consent. Keep the chosen environment/context in the host's
|
|
90
|
+
MCP arguments too. Inspect identity before making changes; an empty project
|
|
91
|
+
list is not a reason to switch accounts or environments automatically. Use the
|
|
92
|
+
same flags for logout when revoking that connection.
|
|
53
93
|
|
|
54
94
|
For a remote machine, use `geniustasker auth login --flow device --no-browser`.
|
|
55
95
|
For Chrome specifically, use `--browser chrome`. `--context work` creates a
|
|
@@ -60,6 +100,14 @@ and approve it, or uncheck “All” to choose individual projects and memories
|
|
|
60
100
|
remove any permissions you do not need. Opening the page grants nothing, and
|
|
61
101
|
existing connections keep their approved permissions.
|
|
62
102
|
|
|
103
|
+
Without `--scope`, browser and device login leave the permission proposal to the
|
|
104
|
+
selected server. A newer CLI can therefore connect before that environment has
|
|
105
|
+
deployed every permission in its bundled command catalogue. The human still
|
|
106
|
+
reviews and approves the visible selection; login never updates an existing
|
|
107
|
+
connection or profile. Features absent from the deployment remain unavailable.
|
|
108
|
+
An explicit `--scope` is sent unchanged: unsupported permissions are rejected,
|
|
109
|
+
not silently removed or retried with different access.
|
|
110
|
+
|
|
63
111
|
To narrow the permissions offered by the browser, pass `--scope`, for example:
|
|
64
112
|
|
|
65
113
|
```sh
|
|
@@ -80,7 +128,7 @@ launching stdio servers with this entry (adjust environment/context as needed):
|
|
|
80
128
|
"mcpServers": {
|
|
81
129
|
"geniustasker": {
|
|
82
130
|
"command": "npx",
|
|
83
|
-
"args": ["-y", "@pulsarinteractive/geniustasker@0.1.0-beta.
|
|
131
|
+
"args": ["-y", "@pulsarinteractive/geniustasker@0.1.0-beta.14", "mcp", "--context", "work"]
|
|
84
132
|
}
|
|
85
133
|
}
|
|
86
134
|
}
|
|
@@ -535,6 +583,15 @@ current authorization and semantic validation remain server-side.
|
|
|
535
583
|
|
|
536
584
|
## Agent runs
|
|
537
585
|
|
|
586
|
+
Task create/update and manual lock input schemas are available through
|
|
587
|
+
`commands describe --kind task.create` (or `task.update`, `task.lock`,
|
|
588
|
+
`task.unlock`, `run.lock`, `run.unlock`). Read
|
|
589
|
+
[task locks and current availability](TASK_LOCKS.md) before a lock or unlock.
|
|
590
|
+
These source additions are not a claim of npm publication or production rollout.
|
|
591
|
+
Use `--task-state v1` on task resource/change/watch reads only after the tasks
|
|
592
|
+
collection advertises `taskState: "v1"` in capabilities. Omission preserves the
|
|
593
|
+
legacy response shape; the client does not silently request the new projection.
|
|
594
|
+
|
|
538
595
|
Discover the `agentRuns` collection and `run.claim`, `run.checkpoint`, `run.patch`,
|
|
539
596
|
`run.finish` commands before using them. Inspect each schema with
|
|
540
597
|
`geniustasker commands describe --kind run.claim --json`, then use the normal
|
|
@@ -550,3 +607,16 @@ Add `--prepare` to journal it without sending. Keep its stable `eventId` and
|
|
|
550
607
|
`tasker_run_event`. A `process_exited` event only requests review, with explicit
|
|
551
608
|
next steps. See [host event fields, examples and recovery](../skills/geniustasker/references/host-events.md)
|
|
552
609
|
before connecting a host callback; no host hook is installed automatically.
|
|
610
|
+
|
|
611
|
+
### Personal focus and temporary deferral
|
|
612
|
+
|
|
613
|
+
`intents read --scope project:ID` returns the connected owner's bounded
|
|
614
|
+
instructions. `intents preview --file selection.json` checks exact tasks, time
|
|
615
|
+
window and overlapping instructions without writing. Follow
|
|
616
|
+
[the planning guide](PLANNING.md#temporary-personal-focus-and-deferral) and
|
|
617
|
+
`workflow show --topic intents`, then use the ordinary immutable
|
|
618
|
+
`intent.change` command flow. Availability is determined by service capabilities.
|
|
619
|
+
|
|
620
|
+
## Feedback
|
|
621
|
+
|
|
622
|
+
Use [the feedback guide](FEEDBACK.md) for authorized bug reports, feature suggestions, screenshots and final decisions. Availability follows current server capabilities, grants and plan. Preparing a report does not send it; submit after the user’s reporting instruction or agreement.
|
package/docs/FEEDBACK.md
ADDED
|
@@ -0,0 +1,96 @@
|
|
|
1
|
+
# Private feedback
|
|
2
|
+
|
|
3
|
+
Feedback reports a problem or suggestion about GeniusTasker itself. It does not
|
|
4
|
+
create a project ticket. Availability depends on the connected deployment,
|
|
5
|
+
account admission, plan and current consent. Installing this source does not
|
|
6
|
+
activate the backend or expand an existing connection.
|
|
7
|
+
|
|
8
|
+
Free accounts can report bugs and recover delivery receipts with `feedback:write`.
|
|
9
|
+
Starter/Pro can also suggest features and read their feedback with `feedback:read`.
|
|
10
|
+
Capabilities reflect both the plan and the current connection/profile grants.
|
|
11
|
+
Start with `feedback capabilities`; do not infer rights from an old receipt.
|
|
12
|
+
If a new grant is needed, the human reconnects in the relevant host. CLI consent
|
|
13
|
+
does not change a ChatGPT connection. Login without `--scope` leaves the proposed
|
|
14
|
+
permissions to that server; an explicit `--scope` remains strict.
|
|
15
|
+
|
|
16
|
+
## Prepare and submit
|
|
17
|
+
|
|
18
|
+
Use the same context and environment throughout. These examples target an
|
|
19
|
+
already connected production context; production remains limited to admitted
|
|
20
|
+
accounts. Supply the title and observed description in a local JSON draft:
|
|
21
|
+
|
|
22
|
+
```json
|
|
23
|
+
{
|
|
24
|
+
"type": "BugReport",
|
|
25
|
+
"category": "sync",
|
|
26
|
+
"title": "The task title stayed unchanged on the other device",
|
|
27
|
+
"description": "The web page showed the saved title. The phone still showed its previous title after reopening the project.",
|
|
28
|
+
"attachments": []
|
|
29
|
+
}
|
|
30
|
+
```
|
|
31
|
+
|
|
32
|
+
```sh
|
|
33
|
+
geniustasker feedback capabilities --environment production --context work --json
|
|
34
|
+
geniustasker feedback prepare --file draft.json --environment production --context work --json
|
|
35
|
+
geniustasker feedback submit --file command.json --environment production --context work --json
|
|
36
|
+
```
|
|
37
|
+
|
|
38
|
+
Save the exact `data.command` returned by prepare as `command.json`. Preparation
|
|
39
|
+
reads the current epoch if omitted and saves a private local journal; it does not
|
|
40
|
+
submit. The command contains `commandId`, `issuedAt`, `draftId`, `expectedEpoch`
|
|
41
|
+
and the complete report. Show the content and selected attachments, then submit
|
|
42
|
+
after agreement. A direct instruction to report already authorizes that report.
|
|
43
|
+
Never silently send every tool error or attach private context automatically.
|
|
44
|
+
|
|
45
|
+
After an uncertain response, use `feedback receipt --id COMMAND_ID` or
|
|
46
|
+
`commands retry --operation COMMAND_ID` with the same context and environment.
|
|
47
|
+
`commands inspect --operation COMMAND_ID` reads the local journal. Retry reuses
|
|
48
|
+
the exact saved body and original connection/profile/session. A confirmed local
|
|
49
|
+
receipt avoids another request; it does not prove current status. Never edit
|
|
50
|
+
the command or generate a fresh ID to retry.
|
|
51
|
+
|
|
52
|
+
## Screenshots
|
|
53
|
+
|
|
54
|
+
Choose one draft UUID and current epoch first. Before submitting, prepare a
|
|
55
|
+
selected local file with `feedback media-prepare --draft-id DRAFT_UUID --epoch
|
|
56
|
+
CURRENT_EPOCH --file screenshot.png`. Save the returned metadata, then use
|
|
57
|
+
`feedback media-open --file upload.json`. Reserve only user-authorized images;
|
|
58
|
+
this step consumes storage admission. Upload with `feedback media-upload
|
|
59
|
+
--operation UPLOAD_UUID --file screenshot.png` and retain the same bytes and ID
|
|
60
|
+
after uncertainty. Confirmed upload replay does not reread the file.
|
|
61
|
+
|
|
62
|
+
Put ready asset IDs in `attachments` and use the same `draftId` when preparing
|
|
63
|
+
the final report. Up to three static PNG/JPEG/WebP images can accompany the
|
|
64
|
+
initial request. Source bytes are capped at 4,000,000; the server normalizes to
|
|
65
|
+
WebP, at most 1080 px on the longest side and 1,000,000 stored bytes. Feedback and
|
|
66
|
+
global account storage quotas both apply. Local source inspection does not
|
|
67
|
+
replace server validation. Text-only reporting remains available if image
|
|
68
|
+
storage is refused. Existing reports cannot receive additional images.
|
|
69
|
+
|
|
70
|
+
`feedback media-download --id ASSET_ID --file new-image.webp` downloads an
|
|
71
|
+
authorized image without overwriting a local file. `feedback media-remove --id
|
|
72
|
+
ASSET_ID` removes only an explicitly selected image at the user's request. Both
|
|
73
|
+
accept the usual context/environment options. Retrying removal uses the same ID.
|
|
74
|
+
|
|
75
|
+
## Decision tracking and MCP
|
|
76
|
+
|
|
77
|
+
`feedback list` accepts `--status`, `--type` and a returned `--after` cursor.
|
|
78
|
+
`feedback show --id REQUEST_ID` reads an exact decision. Read only requested
|
|
79
|
+
pages; do not poll. `feedback mark-read --file read.json` accepts
|
|
80
|
+
`{"id":"REQUEST_UUID","request":{"expectedEpoch":"CURRENT_EPOCH","revision":"VIEWED_REVISION"}}`
|
|
81
|
+
and requires `feedback:write` in addition to plan eligibility.
|
|
82
|
+
|
|
83
|
+
The team records a status, resolution and optional final message. There is no
|
|
84
|
+
conversation or reply command; the `messages` array has at most one final team
|
|
85
|
+
message. “Saved for later” does not promise delivery. A closure can send a push
|
|
86
|
+
according to the user's notification settings; intermediate changes do not.
|
|
87
|
+
Private administrative notes and duplicate targets are not exposed.
|
|
88
|
+
|
|
89
|
+
Local MCP exposes `tasker_feedback_*` tools for the same workflow, plus local
|
|
90
|
+
image preparation/upload/download. Hosted MCP exposes capabilities, prepare,
|
|
91
|
+
submit, list, detail, receipt, mark-read and image removal. Hosted preparation
|
|
92
|
+
returns an immutable object without a local journal: retain and resubmit the
|
|
93
|
+
exact object after uncertainty. It never accepts arbitrary URLs or filesystem
|
|
94
|
+
paths. Hosted ChatGPT attachment import is not implemented; a visible chat
|
|
95
|
+
attachment is not an uploaded feedback image. Use text-only feedback there
|
|
96
|
+
unless an authorized ready asset already exists.
|
package/docs/INTEGRATIONS.md
CHANGED
|
@@ -4,11 +4,31 @@ Choose either direct MCP configuration or the plugin bundle, to avoid registerin
|
|
|
4
4
|
the same server twice. Authenticate once per chosen local context. Permissions
|
|
5
5
|
are selected in the browser and enforced by the service on every request.
|
|
6
6
|
|
|
7
|
+
## Choose the environment
|
|
8
|
+
|
|
9
|
+
The beta CLI and shipped plugin/MCP examples default to development. Production
|
|
10
|
+
access is enabled for admitted accounts and uses the same data as TestFlight.
|
|
11
|
+
For production, use a separate context and include `--environment production`
|
|
12
|
+
in both login and the MCP server arguments. Context names do not select an
|
|
13
|
+
environment; existing development sessions stay separate. Never switch an
|
|
14
|
+
environment or replace a connection just to get around an access refusal.
|
|
15
|
+
|
|
16
|
+
For example, to connect a production context to a direct Codex MCP server:
|
|
17
|
+
|
|
18
|
+
```sh
|
|
19
|
+
npx -y @pulsarinteractive/geniustasker@0.1.0-beta.14 auth login --environment production --context tasker-beta
|
|
20
|
+
codex mcp add geniustasker-production -- npx -y @pulsarinteractive/geniustasker@0.1.0-beta.14 mcp --environment production --context tasker-beta
|
|
21
|
+
```
|
|
22
|
+
|
|
23
|
+
Use either this configuration or the development example below for the intended
|
|
24
|
+
account. If intentionally keeping both, name them distinctly and verify identity
|
|
25
|
+
before working. Adding the production example does not retarget a bundled plugin.
|
|
26
|
+
|
|
7
27
|
## Codex: direct MCP
|
|
8
28
|
|
|
9
29
|
```sh
|
|
10
|
-
npx -y @pulsarinteractive/geniustasker@0.1.0-beta.
|
|
11
|
-
codex mcp add geniustasker -- npx -y @pulsarinteractive/geniustasker@0.1.0-beta.
|
|
30
|
+
npx -y @pulsarinteractive/geniustasker@0.1.0-beta.14 auth login --context work
|
|
31
|
+
codex mcp add geniustasker -- npx -y @pulsarinteractive/geniustasker@0.1.0-beta.14 mcp --context work
|
|
12
32
|
```
|
|
13
33
|
|
|
14
34
|
Restart the host as needed, then inspect its MCP list. Ask the agent to read its
|
|
@@ -39,6 +59,9 @@ without skills, and skills can use the CLI when an MCP host is not configured.
|
|
|
39
59
|
The root `plugin.json` and `mcp.json` follow Agent Plugins 1.0.0. The bundled MCP
|
|
40
60
|
runs locally over stdio and pins a published npm version. Node and npm must be
|
|
41
61
|
available in the execution environment. Authenticate the `work` context first.
|
|
62
|
+
Its shipped arguments target development; for production, use a host-supported
|
|
63
|
+
configuration override or the explicit direct MCP configuration above. Do not
|
|
64
|
+
assume that a production login changes the plugin's environment.
|
|
42
65
|
|
|
43
66
|
The repository also has `.agents/plugins/marketplace.json` for local distribution:
|
|
44
67
|
|
package/docs/PLANNING.md
CHANGED
|
@@ -4,6 +4,53 @@ Use `geniustasker workflow show --topic planning --json` or `tasker_workflow`
|
|
|
4
4
|
with `topic: planning` for the same guide in CLI, local MCP and ChatGPT.
|
|
5
5
|
Inspect `capabilities` and the current record before preparing a command.
|
|
6
6
|
|
|
7
|
+
## Progress and effort
|
|
8
|
+
|
|
9
|
+
Use the task's actual progress: `New` means to do, `InProgress` means work has
|
|
10
|
+
started, and `Done` means the intended work and acceptance criteria are verified.
|
|
11
|
+
Creating a plan, assigning an owner or claiming an agent run does not itself
|
|
12
|
+
start or finish the task. Update progress when the work changes. A finished
|
|
13
|
+
process, published protocol or completed run is not proof that a task is done.
|
|
14
|
+
Never force completion to clear a dependency or make a plan appear finished.
|
|
15
|
+
|
|
16
|
+
`estimateDays` measures approximate effort in working days: `0.1` is one tenth
|
|
17
|
+
of a working day and `1` is one working day. It is not a deadline, elapsed
|
|
18
|
+
calendar time, time spent waiting, or an AI execution timer. Do not convert it
|
|
19
|
+
to hours without the user's working-day convention. Deadlines and protocol
|
|
20
|
+
`dueAfterDays` are separate scheduling choices, never automatic conversions of
|
|
21
|
+
estimated effort.
|
|
22
|
+
|
|
23
|
+
Whenever creating a task or taking charge of existing work, actively check
|
|
24
|
+
`estimateDays`; do not wait for a separate request to consider effort. Supply
|
|
25
|
+
an estimate within the authorized work when the scope and known assumptions
|
|
26
|
+
support one. Apply the same check to each step of a new or explicitly revised
|
|
27
|
+
protocol. Preserve an existing human estimate or deliberate choice unless its
|
|
28
|
+
revision is within the user's request.
|
|
29
|
+
|
|
30
|
+
Use `null` when the effort is unknown. Explain the useful missing assumption in
|
|
31
|
+
the planning response or an authorized progress note, then reassess after that
|
|
32
|
+
uncertainty is clarified. There is no separate estimate-reason API field. Do
|
|
33
|
+
not manufacture a value to fill a task, convert elapsed AI time into effort,
|
|
34
|
+
or silently replace unknown with a default day. A request to reuse a published
|
|
35
|
+
protocol does not authorize rewriting its steps, digest or saved retries.
|
|
36
|
+
When the scope or understanding changes, reassess the effort and explain a
|
|
37
|
+
proposed or authorized revision. Do not silently replace another person's
|
|
38
|
+
estimate with elapsed agent time or a faster model's runtime.
|
|
39
|
+
|
|
40
|
+
The unreleased source accepts finite estimates from `0.1` through `60`, including
|
|
41
|
+
optional protocol-step estimates. This is not a claim that installed npm
|
|
42
|
+
beta.13 or the connected deployment supports those additions. Inspect current
|
|
43
|
+
command help and capabilities before sending them. An older deployment can
|
|
44
|
+
reject the new field or values below `0.25`; explain the limitation and retain
|
|
45
|
+
the intended estimate rather than rounding it or changing transports to bypass
|
|
46
|
+
validation. Unreleased task create/update help now includes generated field
|
|
47
|
+
schemas: creation requires a title and update preserves omitted fields. Native
|
|
48
|
+
permissions, live references and lifecycle rules remain server-authoritative.
|
|
49
|
+
|
|
50
|
+
For a blocked task or an authorized manual lock, read
|
|
51
|
+
[task locks and current availability](TASK_LOCKS.md). Locked is an effective
|
|
52
|
+
availability state, not a fourth canonical progress value.
|
|
53
|
+
|
|
7
54
|
| Operation | Commands | Required access |
|
|
8
55
|
| ----------------- | -------------------------------------------------------------- | ------------------------------------------------------------------------------------------ |
|
|
9
56
|
| New project | `project.create` | `projects:write`, all-resources consent including future projects, available account quota |
|
|
@@ -77,3 +124,51 @@ keep stable page and block IDs. Folder cycles are rejected by the server.
|
|
|
77
124
|
Prepare then execute one immutable operation. Retain its ID, timestamp and body
|
|
78
125
|
for uncertain retries. Read back affected collections after success. Open apps
|
|
79
126
|
receive the same realtime changes as edits made in the native UI.
|
|
127
|
+
|
|
128
|
+
## Temporary personal focus and deferral
|
|
129
|
+
|
|
130
|
+
Read workflow `intents` before using this capability. It is available only when
|
|
131
|
+
`intent.change` appears in the connected deployment's capabilities. It needs
|
|
132
|
+
`tasks:read` for reads/previews, `tasks:write` for commands and the project's
|
|
133
|
+
existing human-controlled agent opt-in.
|
|
134
|
+
|
|
135
|
+
```sh
|
|
136
|
+
geniustasker workflow show --topic intents --json
|
|
137
|
+
geniustasker intents read --scope project:PROJECT_ID --json
|
|
138
|
+
geniustasker intents preview --file preview.json --json
|
|
139
|
+
geniustasker commands describe --kind intent.change --json
|
|
140
|
+
```
|
|
141
|
+
|
|
142
|
+
A preview file contains `scope` and `selection`: kind `focus` or `defer`, its
|
|
143
|
+
objective, 1–20 exact open `taskIds`, `startsAt` (null means commit time), and an
|
|
144
|
+
explicit `endsAt`. Duration and advance scheduling are each bounded to 14 days.
|
|
145
|
+
Focus objectives are `finish`, `unblock`, `consistency`; defer uses `pause`.
|
|
146
|
+
Resolve a user's local timezone before selecting dates. No recursive selector
|
|
147
|
+
implicitly includes descendants.
|
|
148
|
+
|
|
149
|
+
Local MCP exposes `tasker_intents` and `tasker_intent_preview`; the hosted
|
|
150
|
+
connector has the same tools. A preview makes no changes. Review its full
|
|
151
|
+
replacement list: overlap with part of an earlier group retires that whole
|
|
152
|
+
instruction, not just the overlapping task. Ask for clarification when that
|
|
153
|
+
would exceed the requested change.
|
|
154
|
+
|
|
155
|
+
The command's outer `id` is the **connected human's ID** from identity. Its
|
|
156
|
+
`expectedRevision` is the preview's `intentRevision`, including null only for
|
|
157
|
+
an initially absent instruction aggregate. The separate `data.id` is a new
|
|
158
|
+
instruction UUID. Copy `expectedIntentRevision`, `selection`, `sourceVersion`
|
|
159
|
+
and `replaces` exactly from the preview and use `decision: set`. The server
|
|
160
|
+
owns attribution; never include internal action or author fields.
|
|
161
|
+
|
|
162
|
+
For `decision: cancel`, use the existing instruction's UUID, the latest
|
|
163
|
+
aggregate revision in both revision fields, `selection: null`,
|
|
164
|
+
`sourceVersion: null`, and `replaces: []`. It never revives a superseded focus.
|
|
165
|
+
Use the ordinary immutable command journal on CLI/local MCP or the exact
|
|
166
|
+
prepared command with `tasker_edit_tasks` in the hosted connector. Retry the
|
|
167
|
+
same operation after uncertainty. A definitive conflict requires a fresh read
|
|
168
|
+
and preview, not a forced overwrite.
|
|
169
|
+
|
|
170
|
+
Read back through `tasker_intents`; private command receipts deliberately have
|
|
171
|
+
no public changed items. These instructions do not change shared deadlines,
|
|
172
|
+
status, importance, XP or agent execution claims. Reassignment, deletion and
|
|
173
|
+
access changes make affected instructions stale. They belong only to the
|
|
174
|
+
connected owner; another collaborator's instructions are not exposed.
|
|
@@ -0,0 +1,91 @@
|
|
|
1
|
+
# Task locks and current availability
|
|
2
|
+
|
|
3
|
+
These commands and reads are in unreleased source. They are not evidence of an
|
|
4
|
+
npm publication, enabled production deployment or app release. Discover the
|
|
5
|
+
connected service's capabilities and command help before using them. Keep the
|
|
6
|
+
same account, scope and current permission grants.
|
|
7
|
+
|
|
8
|
+
## Progress is separate from availability
|
|
9
|
+
|
|
10
|
+
Canonical task progress stays `New`, `InProgress` or `Done`. Follow
|
|
11
|
+
[progress and effort guidance](PLANNING.md#progress-and-effort); never send
|
|
12
|
+
`status: "Locked"` in `task.update` or `run.patch`.
|
|
13
|
+
|
|
14
|
+
An unfinished task is effectively Locked while it has a manual reason, an
|
|
15
|
+
unfinished live prerequisite, or both. Unlocking manually removes only the
|
|
16
|
+
manual reason. It does not delete dependencies, finish a prerequisite or grant
|
|
17
|
+
permission to start blocked work. When the last cause disappears, the retained
|
|
18
|
+
New/InProgress progress becomes visible. A lock is not progress, a completion,
|
|
19
|
+
an agent run's Waiting state or a private focus/defer instruction.
|
|
20
|
+
|
|
21
|
+
Already Done work remains Done when a prerequisite is reopened. Locking a Done
|
|
22
|
+
task requires an explicitly intended reopen first. Archived unfinished
|
|
23
|
+
prerequisites still block; absence from a page or cache does not establish
|
|
24
|
+
deletion. Do not remove prerequisites or force Done to obtain an available task.
|
|
25
|
+
|
|
26
|
+
## Read the current source
|
|
27
|
+
|
|
28
|
+
First read capabilities. Only when the `collections` entry with `id: "tasks"`
|
|
29
|
+
advertises `taskState: "v1"`, explicitly select `taskState: "v1"` in MCP task
|
|
30
|
+
resource/change/watch reads, or `--task-state v1` with CLI `resources list`,
|
|
31
|
+
`changes list` and `changes watch`. This signal works with read-only task grants;
|
|
32
|
+
permission to lock is not needed for availability reads. Keep the choice on
|
|
33
|
+
continuations and bounded watches. The client never adds it automatically.
|
|
34
|
+
Omit it on older deployments. Without the option the server returns the legacy
|
|
35
|
+
shape with no availability snapshot, preserving published beta.13 compatibility.
|
|
36
|
+
|
|
37
|
+
Task resource entries and task change `upsert` entries can carry a separate
|
|
38
|
+
`taskAvailability` snapshot. It contains `progress`, effective `status`,
|
|
39
|
+
`manualReason`, `dependencyCount`, up to 20 `blockingIds`, `blockersTruncated`,
|
|
40
|
+
`incomplete` and `source: {epoch, authorizationEpoch, head}`. Removal entries
|
|
41
|
+
have no snapshot. Other collections do not expose this task projection.
|
|
42
|
+
|
|
43
|
+
`Unknown`, `incomplete: true` or an absent snapshot is not proof of availability.
|
|
44
|
+
The server bounds dependency resolution to 1000 distinct prerequisites per
|
|
45
|
+
page. For a task needing a current decision, request that exact `tasks` ID
|
|
46
|
+
through the normal bounded resource read; if its source remains incomplete,
|
|
47
|
+
explain the uncertainty. Do not scan the entire project to reconstruct a graph.
|
|
48
|
+
A truncated blocker list is not an exhaustive list of reasons.
|
|
49
|
+
|
|
50
|
+
Use the snapshot's source epoch, authorization epoch and **task-module head**
|
|
51
|
+
for freshness. A prerequisite change invalidates older availability even if
|
|
52
|
+
the dependent task's row revision does not change and no dependent delta is
|
|
53
|
+
sent. The snapshot head can be newer than a change page's historical checkpoint.
|
|
54
|
+
Keep those values separate; never rewrite one from the other. Discard older
|
|
55
|
+
availability decisions when the source changes and reread the relevant task.
|
|
56
|
+
Any advanced task-module head or checkpoint invalidates older availability,
|
|
57
|
+
including a delta containing only `remove` with no dependent snapshot. There
|
|
58
|
+
is no dependent fanout. Before starting or completing work, reread that exact
|
|
59
|
+
task with the advertised opt-in and use its current availability.
|
|
60
|
+
The client revalidates cached resource reads with the server; historical output
|
|
61
|
+
and a command receipt are not a present permission or availability grant.
|
|
62
|
+
|
|
63
|
+
## Apply only the authorized manual change
|
|
64
|
+
|
|
65
|
+
Inspect `commands describe --kind KIND --json` or `tasker_command_help` for the
|
|
66
|
+
exact schema, then use the ordinary immutable prepare/execute workflow:
|
|
67
|
+
|
|
68
|
+
| Command | Envelope target and revision | Data |
|
|
69
|
+
| ------------- | ------------------------------------ | ---------------------------------------------------------------------------------------------- |
|
|
70
|
+
| `task.lock` | Task ID and current task revision | `{"reason":"The required test device is unavailable."}` |
|
|
71
|
+
| `task.unlock` | Task ID and current task revision | `{}` |
|
|
72
|
+
| `run.lock` | Live run ID and current run revision | `{"taskRevision":"CURRENT_TASK_REVISION","reason":"The required test device is unavailable."}` |
|
|
73
|
+
| `run.unlock` | Live run ID and current run revision | `{"taskRevision":"CURRENT_TASK_REVISION"}` |
|
|
74
|
+
|
|
75
|
+
Reasons are required plain text, 1–500 characters and nonempty after trimming.
|
|
76
|
+
Use an observed, useful reason; the example is not a reason to invent a blocker.
|
|
77
|
+
The server owns attribution and time. Do not patch `manualLockReason`, include
|
|
78
|
+
an internal `action`, or use another field to simulate locking.
|
|
79
|
+
|
|
80
|
+
The owner of an active agent claim must use **run.lock/run.unlock**, with both
|
|
81
|
+
current revisions. Direct task commands are not a workaround for a claim,
|
|
82
|
+
expired lease, takeover or revision refusal. Read the run and reconcile current
|
|
83
|
+
authority; never switch profiles to evade it. Read both run and task after a
|
|
84
|
+
successful run command because their revisions can change.
|
|
85
|
+
|
|
86
|
+
Preserve operation ID, issuedAt, target, revisions, reason and full body for
|
|
87
|
+
uncertain retry. A definitive conflict requires fresh reads and a new operation
|
|
88
|
+
only for reconciled intent. A refused start/completion under current lock rules
|
|
89
|
+
does not justify automatic retries; explain the cause and continue independent
|
|
90
|
+
authorized work where useful. Read availability after unlocking before saying
|
|
91
|
+
the task is available. No lock receipt proves completion, XP or push delivery.
|
package/examples/codex.toml
CHANGED
package/examples/mcp.json
CHANGED
package/mcp.json
CHANGED
package/package.json
CHANGED
package/plugin.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
{
|
|
2
2
|
"$schema": "https://agent-plugins.org/schemas/1.0.0/plugin.schema.json",
|
|
3
3
|
"name": "geniustasker",
|
|
4
|
-
"version": "0.1.0-beta.
|
|
4
|
+
"version": "0.1.0-beta.14",
|
|
5
5
|
"description": "Track project work and independent product memories with your own agents.",
|
|
6
6
|
"author": {
|
|
7
7
|
"name": "Pulsar Interactive"
|