@genspark/cli 1.13.1 → 1.14.1
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/README.md +8 -5
- package/dist/client.d.ts.map +1 -1
- package/dist/client.js +4 -2
- package/dist/client.js.map +1 -1
- package/dist/commands/aidrive-operands.d.ts +29 -0
- package/dist/commands/aidrive-operands.d.ts.map +1 -0
- package/dist/commands/aidrive-operands.js +178 -0
- package/dist/commands/aidrive-operands.js.map +1 -0
- package/dist/commands/aidrive-upload.d.ts +149 -0
- package/dist/commands/aidrive-upload.d.ts.map +1 -1
- package/dist/commands/aidrive-upload.js +344 -0
- package/dist/commands/aidrive-upload.js.map +1 -1
- package/dist/commands/design.d.ts +9 -0
- package/dist/commands/design.d.ts.map +1 -1
- package/dist/commands/design.js +12 -2
- package/dist/commands/design.js.map +1 -1
- package/dist/commands/genteam-download.d.ts +3 -1
- package/dist/commands/genteam-download.d.ts.map +1 -1
- package/dist/commands/genteam-download.js +4 -1
- package/dist/commands/genteam-download.js.map +1 -1
- package/dist/commands/telephony-recording.d.ts +34 -0
- package/dist/commands/telephony-recording.d.ts.map +1 -0
- package/dist/commands/telephony-recording.js +75 -0
- package/dist/commands/telephony-recording.js.map +1 -0
- package/dist/config.d.ts +13 -0
- package/dist/config.d.ts.map +1 -1
- package/dist/config.js +18 -0
- package/dist/config.js.map +1 -1
- package/dist/file-path-uploads.d.ts +4 -0
- package/dist/file-path-uploads.d.ts.map +1 -0
- package/dist/file-path-uploads.js +34 -0
- package/dist/file-path-uploads.js.map +1 -0
- package/dist/index.d.ts.map +1 -1
- package/dist/index.js +278 -55
- package/dist/index.js.map +1 -1
- package/dist/types.d.ts +2 -0
- package/dist/types.d.ts.map +1 -1
- package/docs/skills.md +3 -3
- package/package.json +1 -1
- package/skills/gsk-aidrive/SKILL.md +175 -85
- package/skills/gsk-analyze-media/SKILL.md +1 -0
- package/skills/gsk-audio-generation/SKILL.md +2 -1
- package/skills/gsk-audio-processing/SKILL.md +1 -0
- package/skills/gsk-audio-transcribe/SKILL.md +1 -0
- package/skills/gsk-batch-crawl-url-and-answer/SKILL.md +1 -0
- package/skills/gsk-batch-understand-videos/SKILL.md +1 -0
- package/skills/gsk-claw-share-link/SKILL.md +1 -0
- package/skills/gsk-crawler/SKILL.md +1 -0
- package/skills/gsk-create-task/SKILL.md +6 -1
- package/skills/gsk-genteam/SKILL.md +4 -3
- package/skills/gsk-get-model-info/SKILL.md +1 -0
- package/skills/gsk-get-service-url/SKILL.md +1 -0
- package/skills/gsk-github/SKILL.md +1 -0
- package/skills/gsk-gmail/SKILL.md +4 -2
- package/skills/gsk-google-calendar/SKILL.md +1 -0
- package/skills/gsk-google-chat/SKILL.md +2 -1
- package/skills/gsk-google-contacts/SKILL.md +1 -0
- package/skills/gsk-google-docs/SKILL.md +1 -0
- package/skills/gsk-google-drive/SKILL.md +27 -7
- package/skills/gsk-google-sheets/SKILL.md +2 -1
- package/skills/gsk-google-slides/SKILL.md +1 -0
- package/skills/gsk-hub/SKILL.md +1 -0
- package/skills/gsk-image-generation/SKILL.md +1 -0
- package/skills/gsk-image-search/SKILL.md +1 -0
- package/skills/gsk-inbox-contacts/SKILL.md +1 -0
- package/skills/gsk-maps-search/SKILL.md +1 -0
- package/skills/gsk-meeting/SKILL.md +1 -0
- package/skills/gsk-microsoft-teams/SKILL.md +2 -1
- package/skills/gsk-notion/SKILL.md +1 -0
- package/skills/gsk-onedrive/SKILL.md +2 -1
- package/skills/gsk-outlook-calendar/SKILL.md +1 -0
- package/skills/gsk-outlook-contacts/SKILL.md +1 -0
- package/skills/gsk-outlook-email/SKILL.md +4 -2
- package/skills/gsk-phone-call/SKILL.md +138 -0
- package/skills/gsk-salesforce/SKILL.md +1 -0
- package/skills/gsk-sb-brain/SKILL.md +1 -0
- package/skills/gsk-scholar-search/SKILL.md +1 -0
- package/skills/gsk-second-brain/SKILL.md +54 -8
- package/skills/gsk-shared/SKILL.md +4 -2
- package/skills/gsk-sharepoint/SKILL.md +2 -1
- package/skills/gsk-sharepoint-upload/SKILL.md +2 -1
- package/skills/gsk-slack/SKILL.md +4 -2
- package/skills/gsk-social-instagram/SKILL.md +1 -0
- package/skills/gsk-social-reddit/SKILL.md +1 -0
- package/skills/gsk-social-twitter/SKILL.md +1 -0
- package/skills/gsk-stock-price/SKILL.md +1 -0
- package/skills/gsk-summarize-large-document/SKILL.md +1 -0
- package/skills/gsk-task-artifacts/SKILL.md +1 -0
- package/skills/gsk-task-continue/SKILL.md +1 -0
- package/skills/gsk-task-info/SKILL.md +1 -0
- package/skills/gsk-task-status/SKILL.md +1 -0
- package/skills/gsk-task-stop/SKILL.md +1 -0
- package/skills/gsk-understand-images/SKILL.md +1 -0
- package/skills/gsk-video-generation/SKILL.md +4 -1
- package/skills/gsk-vm-email-send/SKILL.md +1 -0
- package/skills/gsk-voice-cloning/SKILL.md +5 -4
- package/skills/gsk-web-search/SKILL.md +1 -0
- package/skills/gsk-youtube/SKILL.md +1 -0
|
@@ -12,6 +12,7 @@ description: Make a real AI phone call on the user's behalf. Validates prerequis
|
|
|
12
12
|
telephony sms send` — same setup, lineage, and disclosure rules.
|
|
13
13
|
metadata:
|
|
14
14
|
category: phone-call
|
|
15
|
+
generated_by: genspark_tool_cli/generate_skills.py
|
|
15
16
|
requires:
|
|
16
17
|
bins:
|
|
17
18
|
- gsk
|
|
@@ -41,6 +42,143 @@ MANDATORY after every send: report to the user the recipient, delivery status, s
|
|
|
41
42
|
gsk telephony sms send [options] # see `gsk telephony sms send --help`
|
|
42
43
|
```
|
|
43
44
|
|
|
45
|
+
## Getting the recording file
|
|
46
|
+
|
|
47
|
+
`gsk telephony call detail <project_id>` returns the call record and a recording link. To hand the audio file itself to the user or a channel, download it:
|
|
48
|
+
|
|
49
|
+
gsk telephony call recording <project_id> ./call.wav
|
|
50
|
+
|
|
51
|
+
Default is dual-channel (L=callee, R=caller). Add `--channels mono` for a single server-mixed track. `--info` reports duration and size without downloading.
|
|
52
|
+
If the answer is `recording_not_ready`, the call has not finished finalising — poll `gsk telephony call status` and retry.
|
|
53
|
+
|
|
54
|
+
## The flow — a phone task from ask to delivered result
|
|
55
|
+
|
|
56
|
+
`dial` is one step of a flow, not the whole job. Run the flow below for every
|
|
57
|
+
phone task; the sections above hold the command details.
|
|
58
|
+
|
|
59
|
+
1. **Settle everything up front, once.** Before the first dial, confirm with
|
|
60
|
+
the user in a single exchange: the goal, the hard constraints, what the call
|
|
61
|
+
may agree to and what it must decline (payment, deposit, contract, private
|
|
62
|
+
data), how many calls are allowed, and the language the callee speaks.
|
|
63
|
+
Everything after this runs without coming back to ask. There is no live
|
|
64
|
+
channel during a call: the voice agent works from `purpose` alone.
|
|
65
|
+
2. **Check readiness.** Run `gsk telephony setup-status` whenever the ask is
|
|
66
|
+
about calling. Not ready → hand the user the returned `setup_url` and stop;
|
|
67
|
+
never dial around a gate. The number and the recipient name come from the user, prior tool results
|
|
68
|
+
or research (a maps search returns the business's phone number) — never
|
|
69
|
+
invented.
|
|
70
|
+
3. **Write a self-sufficient `purpose`.** One string carrying the goal, every
|
|
71
|
+
fact the callee may need, every question that must be answered, what to
|
|
72
|
+
accept and what to decline, and the call language. Nothing can be added
|
|
73
|
+
mid-call; a thin `purpose` is the top cause of a failed call. Use
|
|
74
|
+
`--dry-run` first when the contact is unverified.
|
|
75
|
+
|
|
76
|
+
Every real dial passes a purpose review before anything is dialed
|
|
77
|
+
(`--dry-run` currently stops before it). The review refuses: emergency
|
|
78
|
+
and premium-rate numbers; political purposes; sensitive institutions
|
|
79
|
+
(police non-emergency lines, embassies and consulates, airports,
|
|
80
|
+
military, government security); marketing, sales or any commercial
|
|
81
|
+
solicitation; harassment, scams, threats or other malicious intent;
|
|
82
|
+
call-pumping patterns (silent presence, "never hang up", padding the
|
|
83
|
+
duration); any instruction to deny being an AI, impersonate a real
|
|
84
|
+
person or hide who the call is for; and a purpose the callee cannot act
|
|
85
|
+
on (a reservation without date, time and party size; a scheduling call
|
|
86
|
+
without the caller's availability). Separately, some countries and
|
|
87
|
+
number prefixes are refused as outbound destinations, and a number
|
|
88
|
+
called too often in the last 24 hours is refused. Never write a purpose
|
|
89
|
+
that argues around any of this.
|
|
90
|
+
4. **Dial once, asynchronously, then poll.**
|
|
91
|
+
`gsk telephony call dial <recipient> -c <contact> -p <purpose> --async`
|
|
92
|
+
returns at once with `project_id`. Poll `gsk telephony call status
|
|
93
|
+
<project_id>` until the status is terminal. A call can run several minutes
|
|
94
|
+
through IVR and hold; a long call still in progress is not a stuck call. If
|
|
95
|
+
a blocking dial timed out, the call is still live: take the project id from
|
|
96
|
+
the `Dialing... (project: ...)` line and poll its status. **Never dial the
|
|
97
|
+
same number again while a call is not terminal.**
|
|
98
|
+
|
|
99
|
+
When a dial does not end in a conversation, read `call_status` together
|
|
100
|
+
with `dial_attempts` (present only when a dial was actually placed) and
|
|
101
|
+
the reason text in `call_summary`, then pick the class below. Phone
|
|
102
|
+
manners apply throughout: a line that was engaged, unanswered or refused
|
|
103
|
+
is not chased, and the system accepts at most 4 dials to one number per
|
|
104
|
+
user per day (6 across all users) — plan every retry inside that.
|
|
105
|
+
|
|
106
|
+
- `blocked` or `dial_rejection_reason: purpose_rejected` — the purpose
|
|
107
|
+
review refused it; nothing was dialed.
|
|
108
|
+
- It asks for a fact you already hold from step 1 (date, time, party
|
|
109
|
+
size, your availability, who is calling): add it to `purpose` and
|
|
110
|
+
dial once more, no question to the user.
|
|
111
|
+
- "Failed to analyze the purpose of the call": the review did not
|
|
112
|
+
run; dial once more.
|
|
113
|
+
- It names information you do not have: ask the user for exactly
|
|
114
|
+
that, then dial.
|
|
115
|
+
- It names a policy category (political, marketing, malicious, call
|
|
116
|
+
pumping, hiding the AI or the principal): tell the user which one
|
|
117
|
+
and quote the reason; dial again only if the user changes the goal
|
|
118
|
+
itself — rewording the same intent is refused again.
|
|
119
|
+
- The purpose does not fit the callee: confirm the target with the
|
|
120
|
+
user.
|
|
121
|
+
- `failed` with no `dial_attempts` — a check refused it before any dial.
|
|
122
|
+
- "called too frequently": retry later. The limit is per 24 hours and
|
|
123
|
+
the message carries no timer; tell the user this callee cannot be
|
|
124
|
+
reached before tomorrow.
|
|
125
|
+
- Destination refused (country or region not supported, number
|
|
126
|
+
blocked): do not retry — final for that party. Never try another
|
|
127
|
+
number for the same callee to get around it.
|
|
128
|
+
- `code=setup_incomplete`, a rejected caller-ID choice, a membership
|
|
129
|
+
or credit limit: hand the user the `setup_url` or the message, wait,
|
|
130
|
+
and run `setup-status` again before dialing.
|
|
131
|
+
- Any other reason ("unexpected error", "temporarily unavailable"):
|
|
132
|
+
one retry a few minutes later; if it fails again, stop and report.
|
|
133
|
+
- `busy` — the line was engaged. Do not redial on your own; report it.
|
|
134
|
+
If the user wants another try, wait at least 30 minutes.
|
|
135
|
+
- `no-answer` — it rang out. At most one more try within the count
|
|
136
|
+
agreed in step 1, at least 30 minutes later and inside the callee's
|
|
137
|
+
daytime (never before 08:00 or after 21:00 callee local time); a
|
|
138
|
+
second no-answer ends the task for that party — report.
|
|
139
|
+
- `canceled` — stopped from our side (a hang-up or a timeout before
|
|
140
|
+
pickup). Do not redial unless the user asks.
|
|
141
|
+
- `failed` with `dial_attempts` — the carrier could not complete the
|
|
142
|
+
call. Read `sip_response_code`, `error_code`, `error_message` and
|
|
143
|
+
`dial_rejection_reason` from the dial result or status when present.
|
|
144
|
+
`dial_attempts: 2` means the system already retried; do not add an
|
|
145
|
+
automatic retry. SIP 404, 410 or 484: ask the user to verify the
|
|
146
|
+
number, never redial the unchanged number. SIP 403 or a destination
|
|
147
|
+
rejection: stop; do not change numbers to bypass it. SIP 603: treat
|
|
148
|
+
it as a refusal, no redial that day. For other or missing codes,
|
|
149
|
+
report the available reason without guessing; retry only after the
|
|
150
|
+
user confirms, at least 10 minutes later. Older servers may omit
|
|
151
|
+
these fields; do not infer a cause from their absence.
|
|
152
|
+
- `completed` — judge from the transcript (step 6). Voicemail or an
|
|
153
|
+
unattended IVR counts as not reached: at most one more try, not
|
|
154
|
+
within the same hour, and never a second voicemail on the same day.
|
|
155
|
+
|
|
156
|
+
Whatever the class: never split a refused goal into several smaller
|
|
157
|
+
calls, and never change the contact to make a refused purpose pass.
|
|
158
|
+
5. **Fetch the record of each call.** On terminal status,
|
|
159
|
+
`gsk telephony call detail <project_id>` returns the summary and the full
|
|
160
|
+
transcript; `gsk telephony call recording <project_id> ./call.wav` saves
|
|
161
|
+
the audio file (see "Getting the recording file" above). Redirect stdout
|
|
162
|
+
to a file and parse the file: long transcripts truncate in context.
|
|
163
|
+
6. **Judge every `completed` call from the transcript, not from the
|
|
164
|
+
status.** `completed` includes voicemail, refusals and "we will call you
|
|
165
|
+
back". Goal met → the task ends; do not dial again to "confirm".
|
|
166
|
+
Terminal (declined, closed, will not talk to an AI, goal impossible) →
|
|
167
|
+
stop, do not redial. Decision needed (payment, identity, legal) → the
|
|
168
|
+
call declined as told by `purpose`; raise it to the user now, between
|
|
169
|
+
calls. Not reached (voicemail, IVR loop, an endless hold) → the retry
|
|
170
|
+
table in step 4 applies. Any status other than `completed` is handled by
|
|
171
|
+
that table too.
|
|
172
|
+
7. **Deliver per call.** Report how many calls were made and, for each: whom
|
|
173
|
+
you called, when, status, real duration, what was agreed (quoted from the
|
|
174
|
+
transcript), the transcript or where it is (`detail`), and the recording
|
|
175
|
+
file. Report in the user's language; the call itself runs in the callee's.
|
|
176
|
+
|
|
177
|
+
Keep one line per call in a durable place as you go (`project_id`, status,
|
|
178
|
+
goal met, one-line summary) and build the report from it. Several candidates:
|
|
179
|
+
call them one at a time and stop at the first acceptable result; for a
|
|
180
|
+
"compare all" goal, reach every candidate with the same `purpose` first.
|
|
181
|
+
|
|
44
182
|
## Flags
|
|
45
183
|
|
|
46
184
|
| Flag | Required | Description |
|
|
@@ -90,11 +90,12 @@ Decide where the thing lives, then use that door once:
|
|
|
90
90
|
| The user asks about... | Do this |
|
|
91
91
|
|---|---|
|
|
92
92
|
| Something they knew, wrote, received or decided | `gsk sb-brain grep -q <short keyword>` across sources, or `-s <source>` when they named one. Then `read -s <hit.source> --repo_id <hit.repo_id> -p <hit.path>` the best hit. `grep` is a fixed string, case-insensitive; not a regex |
|
|
93
|
-
| Themselves, their preferences,
|
|
94
|
-
| A
|
|
95
|
-
| A
|
|
96
|
-
|
|
|
97
|
-
|
|
|
93
|
+
| Themselves, their preferences, their work | `read -s memo -p System/memory/Memory.md` first |
|
|
94
|
+
| A known Drive file | `gsk drive read --id <entry>` or `read --path <path>`; preserve a selected `--source`. Discover only an unknown filename with `find --scope all -q <name>` |
|
|
95
|
+
| A topic inside a selected Drive folder | `gsk drive search --query <topic> --query_scope subtree` with the selected `--id`, `--path` or `--source` |
|
|
96
|
+
| A Drive folder's contents | `gsk drive tree` with the selected address; existing overviews help navigation |
|
|
97
|
+
| An unknown Team Drive or shared source | `ls --workspace shared_drive` or `ls --workspace shared_with_me`, then reuse its `source_id` with `read`/`search`/`tree` |
|
|
98
|
+
| A pasted Drive link or share card | `gsk drive read --url <link>` for files, `gsk drive tree --url <link>` for folders; the resolver follows live shortcuts |
|
|
98
99
|
| A document card in a GenTeam conversation | `de attachment-view --ref "<the card row's attachment_ref>" --target <channel>` on any member's turn while its share is open to the conversation; when that refuses, the three steps in §4 on your creator's turn: the link from the message text, `gsk drive ls --url <link>`, then `get_readable_url` (or `download` for Office and PDF) |
|
|
99
100
|
| A folder card in a GenTeam conversation | `de attachment-list --ref "<the card row's attachment_ref>" --target <channel>` lists its children (`--entry <folder id>` for a sub-folder, `--after <entry id>` for the next page); `de attachment-view --ref "<the card ref>" --entry <child id> --target <channel>` reads one child — both on any member's turn while its share is open to the conversation |
|
|
100
101
|
| An attachment in the current conversation | The conversation's own attachment path, not the Drive |
|
|
@@ -102,6 +103,49 @@ Decide where the thing lives, then use that door once:
|
|
|
102
103
|
|
|
103
104
|
Always `read` or fetch a file before quoting or acting on it. If nothing relevant comes back, try one variant (shorter prefix, nickname, the user's other language), then say what you covered. Conflicting sources: surface both with dates.
|
|
104
105
|
|
|
106
|
+
For a topical question, stop collecting once original evidence covers the
|
|
107
|
+
requested facts. Read another file or range to close a named gap or resolve a
|
|
108
|
+
conflict, not just for completeness. A request for a full-document summary still
|
|
109
|
+
requires every range. Do not expand the source root again after locating the
|
|
110
|
+
relevant files unless the evidence is insufficient.
|
|
111
|
+
|
|
112
|
+
Batch `read --ids` accepts at most **6 files**, sharing a **16 KB text budget**
|
|
113
|
+
(with response metadata bounded separately); `char_limit` is not a promise of
|
|
114
|
+
24 KB per file or per batch. Use small batches and follow each file's own
|
|
115
|
+
`next_read`. Pass these limits and source identifiers to delegated readers too.
|
|
116
|
+
|
|
117
|
+
Choose the next evidence-gathering or editing step once and execute it. Revisit
|
|
118
|
+
the plan only when a result, constraint or user request changes it. Reuse facts,
|
|
119
|
+
file identifiers and guides already available in this conversation.
|
|
120
|
+
For an overview of a large folder, track which file groups support each topic
|
|
121
|
+
and fill uncovered groups before claiming complete coverage. Reading selected
|
|
122
|
+
concept summaries alone is sampling, not reading every file in the folder.
|
|
123
|
+
|
|
124
|
+
For Drive, follow `../gsk-aidrive/SKILL.md` for read ranges, search cursors and
|
|
125
|
+
coverage. Existing overview helps choose vocabulary and a starting folder; it
|
|
126
|
+
is not source evidence and must not exclude relevant sibling folders. Missing
|
|
127
|
+
or stale overview does not block search. Request up to 6 useful overviews with `--ids` per call; missing/stale summaries enqueue asynchronously. Continue search/read without polling. Use summaries to plan the next search/read, then read original text for exact facts or uncovered details. Folder summaries can use names and child summaries; they do not prove every descendant was read.
|
|
128
|
+
`coverage=complete`/`source_complete` describe the index; `contains_full_document`
|
|
129
|
+
describes this response. A full summary needs all ranges from offset 0: follow
|
|
130
|
+
the returned `next_read` fields as flags to `gsk drive read` until exhausted and
|
|
131
|
+
`source_complete` is true. On older servers, use `--id` with `next_offset` and
|
|
132
|
+
`source_version`; do not combine multi-file `--ids` with a shared offset.
|
|
133
|
+
Use `--char_limit`/`--limit`/`--byte_limit` to bound output; do not truncate JSON
|
|
134
|
+
with `head` and lose coverage or continuation. Context query results are directly
|
|
135
|
+
in `data`; `get_readable_url` uses `session_state.aidrive_result.readable_url`.
|
|
136
|
+
For `index_pending`, reuse `retry_read` as flags to `gsk drive read` once; it waits
|
|
137
|
+
within a bounded server budget, so do not add a shell sleep. Use `gsk --refresh`
|
|
138
|
+
for that retry the first time you see `wait_seconds`, so a cached CLI schema
|
|
139
|
+
recognizes the new flag. Older servers return
|
|
140
|
+
only `retry_after_seconds`: wait that interval and retry once. If still unavailable
|
|
141
|
+
or incomplete, use `get_readable_url --id <entry>` or `download` and an appropriate
|
|
142
|
+
parser, stating the actual coverage. Use `--id`, not restore's `--entry_id`.
|
|
143
|
+
If a new action/flag is rejected, retry once with `gsk --refresh`; an older server
|
|
144
|
+
may still need the byte-read fallback. Active Office content uses `wo-peer` (§7).
|
|
145
|
+
Permission refusals are not a reason to try another read lane.
|
|
146
|
+
Cite `[title](returned url)` verbatim for Drive evidence. Never prefix that
|
|
147
|
+
URL with `memo:`, which opens a vault note instead of the Drive source.
|
|
148
|
+
|
|
105
149
|
## 6. Writing
|
|
106
150
|
|
|
107
151
|
Decide WHERE before you write, in this order:
|
|
@@ -115,8 +159,8 @@ Commands:
|
|
|
115
159
|
- Notes: `gsk sb-git commit` — full file content, `expected_parent` set.
|
|
116
160
|
- New My Drive files: `gsk drive upload --local_file <file> --upload_path <path> --on-conflict error`. Never keep the `/My Drive` spelling in the path; that literal would create a `My Drive/` folder. If the path exists, resolve whether the user meant that existing entry (§7) or a genuinely new file with a distinct name. Do not retry with overwrite or let automatic renaming turn an update into a duplicate.
|
|
117
161
|
- Team Drive or a shared source: add `--workspace shared_drive|shared_with_me --source <source_id>`. The source drive owns the bytes and quota. GenTeam channel drives take files at the root only.
|
|
118
|
-
- `upload` replies carry the new entry's `url` and `item.id`;
|
|
119
|
-
-
|
|
162
|
+
- `upload` replies carry the new entry's `url` and `item.id`; return that URL directly. A `read` result's `url` is also ready to cite. Do not run `gsk me`, `find`, or reconstruct a link when the result already supplies its URL. Only on an older runtime whose reply lacks it, recover the id with `find -q <name>` in the destination.
|
|
163
|
+
- Only when a URL is missing, the `<drive>` part of a link is the user's own id for My Drive — reuse a known owner; otherwise `gsk me` returns it as `cogen_id`. For a Team Drive or shared source it is the `drive_id` on that source's row (the part of `source_id` before the colon).
|
|
120
164
|
|
|
121
165
|
**The link you hand back.** For anything in Drive, the only link for people is the entry's share link:
|
|
122
166
|
|
|
@@ -126,6 +170,8 @@ Commands:
|
|
|
126
170
|
|
|
127
171
|
When a `gsk drive` reply already carries one, hand that `url` back as it is: `share`, `access` and `link` replies print `Link: ...`, and shortcut rows in `ls`/`find` show one. It may start with `/aidrive/s/` instead of `/second-brain/s/` — the same landing for an account whose Second Brain drive surface is not rolled out yet; both forms open, so never rewrite one into the other. Build the link yourself only when no reply gave you one (a file you just uploaded, a `find` hit), from the `<drive>` rule above and the row's `id`.
|
|
128
172
|
|
|
173
|
+
A reply's `Open folder` / `browse_url` opens the containing folder by the same entry-link scheme; use it only when the user wants the folder. For an uploaded or read file, prefer its `Link` / `url`. Only the caller's Drive home uses a home-page URL. Do not reconstruct folder links with `?path=` or file links with `#p_fid=`.
|
|
174
|
+
|
|
129
175
|
Never hand out a `get_readable_url` (a signed download that opens nothing), a `/aidrive/files/...` folder page, or an editor route (`/second-brain/doc?id=...` is an in-app destination, not a share link). For Notes, hand back the `memo:` link.
|
|
130
176
|
|
|
131
177
|
**Handing a file to a GenTeam conversation.** Choose one delivery for each result, within the current turn's permissions:
|
|
@@ -164,7 +210,7 @@ Then one round: one focused `grep` (or one `find` for files) and one `read` of t
|
|
|
164
210
|
- A PDF is produced by exporting a paginated web document (`gsk doc export -f pdf`) and uploading it; PDF content cannot be edited in place from an agent. `download_file --file_url` can import one.
|
|
165
211
|
- `upload` replies carry `url` and `item.id` since 2026-09-15; on an older runtime recover the id with `find`.
|
|
166
212
|
- `doc-search` indexes My Drive only. It is switched on for internal accounts only.
|
|
167
|
-
- `gsk drive`
|
|
213
|
+
- `gsk drive read` returns persisted indexed text, not the live Office room. Pending or partial extraction is not complete evidence; use the reading guidance in §5. Byte downloads remain subject to the source owner's download policy.
|
|
168
214
|
- In GenTeam, `gsk drive` is creator-turn only, DM roots are discoverable without granting access to referenced sources, and `wo-peer` is on PATH only when your creator is enabled for it (`gk_sb_collab_peer`) — without it, Office documents cannot be changed from there. Since 2026-09-18 a member's turn may change a card shared editable to the conversation through the room (the room acts as that member); everything else in the room stays creator-turn only. A document shared into a DM grants the person, not the conversation; the recipient finds it under `shared_with_me`.
|
|
169
215
|
|
|
170
216
|
Each line disappears as the platform closes it. If a limit below is gone, this skill is stale — say so.
|
|
@@ -213,13 +213,15 @@ The `gsk drive upload` command supports streaming local files directly to AI Dri
|
|
|
213
213
|
# Upload a local file to AI Drive (streaming, supports 100MB+ files)
|
|
214
214
|
gsk drive upload --local_file ./report.pdf --upload_path /docs/report.pdf
|
|
215
215
|
gsk drive upload --local_file ./photo.png # defaults to /photo.png
|
|
216
|
-
gsk drive upload --local_file ./doc.pdf --upload_path /docs/doc.pdf --
|
|
216
|
+
gsk drive upload --local_file ./doc.pdf --upload_path /docs/doc.pdf --overwrite true # edit in place: same id, same link
|
|
217
|
+
gsk drive upload --local_file ./report.pdf --upload_path /docs/report.pdf --to ann@example.com # upload and share; reply carries the link
|
|
217
218
|
```
|
|
218
219
|
|
|
219
220
|
| Flag | Description |
|
|
220
221
|
|------|-------------|
|
|
221
222
|
| `--local_file <path>` | Local file path to upload (streaming, no size limit) |
|
|
222
|
-
| `--
|
|
223
|
+
| `--overwrite true` | Replace the existing file at the destination path in place; the file keeps its id and share link (`--override` is the older spelling) |
|
|
224
|
+
| `--to`, `--general_access` | Share the uploaded file right away (My Drive); the reply's `url` opens for those people |
|
|
223
225
|
|
|
224
226
|
## Connectors (integrations)
|
|
225
227
|
|
|
@@ -5,6 +5,7 @@ description: 'SharePoint site, list, and file operations. Actions: list, list_it
|
|
|
5
5
|
search, read_content, read_file, upload.'
|
|
6
6
|
metadata:
|
|
7
7
|
category: general
|
|
8
|
+
generated_by: genspark_tool_cli/generate_skills.py
|
|
8
9
|
requires:
|
|
9
10
|
bins:
|
|
10
11
|
- gsk
|
|
@@ -46,7 +47,7 @@ gsk sharepoint [options]
|
|
|
46
47
|
| `--list_item_id` | No | [read_content] List item ID from SharePoint search results siteInfo. Required for listItem content type. (string) |
|
|
47
48
|
| `--file_id` | No | [read_file] The ID of the file OR the SharePoint URL to read from Microsoft OneDrive/SharePoint/Teams. Can be either a file ID (like '01W56PINYRO3...') or a full SharePoint URL (like 'https://domain.sharepoint.com/...'). (string) |
|
|
48
49
|
| `--raw_content` | No | [read_file] Return the complete converted document text (markdown) without question-answering. For programmatic consumers that need full fidelity; ignores 'question'. (boolean) |
|
|
49
|
-
| `--file_path` | No | [upload]
|
|
50
|
+
| `--file_path` | No | [upload] A Genspark file-wrapper URL, an AI Drive path (aidrive://, /aidrive/ or /mnt/aidrive/), or a local file uploaded automatically by the CLI. Direct API callers must upload local files first. (string) |
|
|
50
51
|
| `--content` | No | [upload] Text content to upload as the file body (encoded as UTF-8). Useful for creating .txt/.md/.json/.csv files without a local copy. Mutually exclusive with ``file_path``. (string) |
|
|
51
52
|
| `--file_name` | No | [upload] File name to create on SharePoint, including the extension (e.g. 'report.pdf'). Required when ``content`` is used; if omitted with ``file_path``, the basename of the local file is used. (string) |
|
|
52
53
|
| `--mime_type` | No | [upload] MIME type of the file. Auto-detected from the file extension if omitted. (string) |
|
|
@@ -4,6 +4,7 @@ version: 1.0.0
|
|
|
4
4
|
description: Upload a file to a SharePoint document library.
|
|
5
5
|
metadata:
|
|
6
6
|
category: sharepoint
|
|
7
|
+
generated_by: genspark_tool_cli/generate_skills.py
|
|
7
8
|
requires:
|
|
8
9
|
bins:
|
|
9
10
|
- gsk
|
|
@@ -28,7 +29,7 @@ gsk sharepoint upload [options]
|
|
|
28
29
|
|
|
29
30
|
| Flag | Required | Description |
|
|
30
31
|
|------|----------|-------------|
|
|
31
|
-
| `<file_path>` (positional) | No |
|
|
32
|
+
| `<file_path>` (positional) | No | A Genspark file-wrapper URL, an AI Drive path (aidrive://, /aidrive/ or /mnt/aidrive/), or a local file uploaded automatically by the CLI. Direct API callers must upload local files first. (string) |
|
|
32
33
|
| `--content` | No | Text content to upload as the file body (encoded as UTF-8). Useful for creating .txt/.md/.json/.csv files without a local copy. Mutually exclusive with ``file_path``. (string) |
|
|
33
34
|
| `-n`, `--file_name` | No | File name to create on SharePoint, including the extension (e.g. 'report.pdf'). Required when ``content`` is used; if omitted with ``file_path``, the basename of the local file is used. (string) |
|
|
34
35
|
| `-u`, `--sharepoint_url` | No | Target folder URL on SharePoint. Example: 'https://company.sharepoint.com/sites/Team/Shared Documents/Reports'. The tool resolves this to a drive + folder. (string) |
|
|
@@ -5,6 +5,7 @@ description: 'Slack messaging operations. Actions: send, search, search_files, r
|
|
|
5
5
|
download_file, lookup, react, upload, update, delete, permalink.'
|
|
6
6
|
metadata:
|
|
7
7
|
category: general
|
|
8
|
+
generated_by: genspark_tool_cli/generate_skills.py
|
|
8
9
|
requires:
|
|
9
10
|
bins:
|
|
10
11
|
- gsk
|
|
@@ -35,6 +36,7 @@ gsk slack [options]
|
|
|
35
36
|
| `--title` | No | [send] Optional title for the message. If provided, creates a rich formatted message with a header. \| [upload] Optional title shown on the Slack file card. Defaults to the file name. (string) |
|
|
36
37
|
| `--fields` | No | [send] Optional array of field objects with 'title' and 'value' keys to create a structured message layout. (array) |
|
|
37
38
|
| `--thread_ts` | No | [send] Optional thread timestamp to reply in a thread. \| [read] Timestamp ('ts') of a thread's parent message. When set, returns that thread's replies instead of the channel timeline. \| [upload] Optional thread timestamp — share the file as a reply in this thread instead of the conversation root. (string) |
|
|
39
|
+
| `--workspace` | No | Optional. Which linked Slack workspace to act in: its team id (e.g. 'T0123ABCD') or its workspace name. Omit to use the default (most recently connected) workspace. The lookup action with lookup_type='workspaces' lists the choices. (string) |
|
|
38
40
|
| `--query` | No | [search] The search query. You can use modifiers like 'in:#channel', 'from:@user', 'has:link', 'before:yyyy-mm-dd'. \| [search_files] Search terms; Slack modifiers such as 'in:#channel', 'from:@user', 'before:yyyy-mm-dd' are honored. (string) |
|
|
39
41
|
| `--count` | No | [search] The maximum number of messages to return. Default is 100. \| [search_files] Maximum files to return (default 20, max 100). (integer) |
|
|
40
42
|
| `--sort` | No | [search] Sort order of results. 'score' for relevance or 'timestamp' for time. Default is 'score'. \| [search_files] 'score' (relevance, default) or 'timestamp' (newest first). (string) |
|
|
@@ -43,7 +45,7 @@ gsk slack [options]
|
|
|
43
45
|
| `--limit` | No | [read] Maximum messages to return (default 20, max 100). \| [lookup] Maximum number of results to return. Default: 50 (integer) |
|
|
44
46
|
| `--cursor` | No | [read] Pagination cursor from a previous call's next_cursor to fetch older messages. (string) |
|
|
45
47
|
| `--file_id` | No | [download_file] Slack file id (starts with 'F'), from a message's files list (read action) or a search_files hit. (string) |
|
|
46
|
-
| `--lookup_type` | No | [lookup] What to look up: 'users' for team members, 'channels' for channels/conversations, 'all' for both. (string, one of: users, channels, all) |
|
|
48
|
+
| `--lookup_type` | No | [lookup] What to look up: 'users' for team members, 'channels' for channels/conversations, 'all' for both, 'workspaces' for the Slack workspaces linked to this account (their team_id feeds the 'workspace' parameter). (string, one of: users, channels, all, workspaces) |
|
|
47
49
|
| `--search_query` | No | [lookup] Optional search query to filter results by name. Case-insensitive partial match on name, display name, or real name. (string) |
|
|
48
50
|
| `--include_bots` | No | [lookup] Whether to include bot users in results. Default: false (boolean) |
|
|
49
51
|
| `--timestamp` | No | [react] The message's 'ts' value (e.g. '1752600000.000100'). \| [update] The message's 'ts' value (e.g. '1752600000.000100'). Must be a message the connected user authored. \| [delete] The message's 'ts' value (e.g. '1752600000.000100'). Must be a message the connected user authored. \| [permalink] The message's 'ts' value (e.g. '1752600000.000100'). (string) |
|
|
@@ -51,7 +53,7 @@ gsk slack [options]
|
|
|
51
53
|
| `--remove` | No | [react] Set true to remove your existing reaction of this emoji instead of adding it. Default: false. (boolean) |
|
|
52
54
|
| `--user_ids` | No | [invite] Slack user IDs to invite (e.g. ['U0123ABCD']). Up to 30 per call. (array) |
|
|
53
55
|
| `--content` | No | [upload] Text content to upload as a file (UTF-8). Provide exactly one of 'content' or 'file_path'. (string) |
|
|
54
|
-
| `--file_path` | No | [upload]
|
|
56
|
+
| `--file_path` | No | [upload] A Genspark file-wrapper URL, an AI Drive path (aidrive://, /aidrive/ or /mnt/aidrive/), or a local file uploaded automatically by the CLI. Direct API callers must upload local files first. (string) |
|
|
55
57
|
| `--file_name` | No | [upload] File name shown in Slack (include the extension, e.g. 'report.pdf'). Required with 'content'; defaults to the path basename for 'file_path'. (string) |
|
|
56
58
|
| `--initial_comment` | No | [upload] Optional message text posted together with the file. (string) |
|
|
57
59
|
| `--text` | No | [update] The full replacement text (mrkdwn) — Slack replaces the whole message body, not a diff. (string) |
|
|
@@ -6,6 +6,7 @@ description: 'Read a task''s execution output: recent activity (messages includi
|
|
|
6
6
|
Use `task status` for cheap polling; use this when you need the content.'
|
|
7
7
|
metadata:
|
|
8
8
|
category: general
|
|
9
|
+
generated_by: genspark_tool_cli/generate_skills.py
|
|
9
10
|
requires:
|
|
10
11
|
bins:
|
|
11
12
|
- gsk
|
|
@@ -5,6 +5,7 @@ description: Generate videos using AI models. Supports text-to-video and image-t
|
|
|
5
5
|
generation.
|
|
6
6
|
metadata:
|
|
7
7
|
category: general
|
|
8
|
+
generated_by: genspark_tool_cli/generate_skills.py
|
|
8
9
|
requires:
|
|
9
10
|
bins:
|
|
10
11
|
- gsk
|
|
@@ -29,7 +30,7 @@ gsk video [options]
|
|
|
29
30
|
|
|
30
31
|
| Flag | Required | Description |
|
|
31
32
|
|------|----------|-------------|
|
|
32
|
-
| `<query>` (positional) | No | Detailed, self-contained description of the video to generate, written in English. This text is sent to the video model verbatim as the generation prompt, so put all the visual detail here — but NEVER output settings: text like '16:9' or '4 seconds' in the prompt may be rendered literally into the scene. Pass aspect ratio via the `aspect_ratio` parameter and length via `duration`. duration is recommended to be 5 ~ 10 seconds. For image-to-image transitions (user provides start and end frames) e.g., 'A timelapse of a flower blooming in a garden'. Omit when passing the prompt by `query_file` instead. (string) |
|
|
33
|
+
| `<query>` (positional) | No | Detailed, self-contained description of the video to generate, written in English. This text is sent to the video model verbatim as the generation prompt, so put all the visual detail here — but NEVER output settings: text like '16:9' or '4 seconds' in the prompt may be rendered literally into the scene. Pass aspect ratio via the `aspect_ratio` parameter and length via `duration`. duration is recommended to be 5 ~ 10 seconds. For image-to-image transitions (user provides start and end frames) e.g., 'A timelapse of a flower blooming in a garden'. Omit when passing the prompt by `query_file` instead. Or omit on a draft_task_id finalize (it reuses the draft's prompt). (string) |
|
|
33
34
|
| `--query_file` | No | Optional. Repo-relative path to a UTF-8 text file containing the prompt, used verbatim in place of `query` (lets you pass a long prompt by path instead of inlining it). Requires `repo_id`. When set, `query` may be omitted. (string) |
|
|
34
35
|
| `--file_name` | No | The name of the video to generate. (string, null) |
|
|
35
36
|
| `-m`, `--model` | Yes | The model to use for video generation. kling/v3: Latest Kling V3. Supports native audio, but generates a SILENT video unless audio_enable=true. image_urls are frames, not references: 1 image = first frame (i2v); 2 images = first + last frame (the 2nd is the END frame). No reference mode (use kling/o3 with reference_mode for reference-driven generation). Pro/Standard/Turbo tiers (set via the tier parameter; turbo is the fastest arm but has no audio and no first/last-frame control). Supported ratios: 16:9, 9:16, 1:1. Duration: 3-15s. kling/o3: Kling O3. REQUIRES at least one input image or a video_url (no text-to-video). image_urls are frames (1 = first, 2 = first + last) unless reference_mode=true: style/character references, max 4 total including elements. video_url switches to video-to-video (video_mode 'reference' = motion guide, 'edit' = restyle in footage). Native audio via audio_enable. Pro/Standard tiers. Supported ratios: 16:9, 9:16, 1:1. Duration: 3-15s. gemini/veo3.1: Gemini Veo 3.1 - Latest version with enhanced quality and features. Supports text-to-video and image-to-video generation with improved quality. image_urls are frames unless reference_mode: 1 image = first frame, exactly 2 = first/last frame transition, 3 (or reference_mode=true) = style/character references (max 3, 8s only). video_url (MP4 source) = video EXTENSION mode: continues from the last frame, 7s, up to 1080p. Duration: 4s/6s/8s. Supported ratios: 16:9, 9:16. Supports tier (standard \| fast). Resolution: 720p, 1080p, 4k. minimax/h3: MiniMax H3 2K video model. Supports text-to-video, image-to-video with first/last frame control, and reference-to-video (up to 9 reference images, 3 reference videos, 3 reference audio clips — refer to them in the prompt as 'Image 1', 'Video 1', 'Audio 1', not @-style). Keyframes and references are mutually exclusive. Supported ratios: adaptive, 21:9, 16:9, 4:3, 1:1, 3:4, 9:16. Duration: 5-15s. Resolution: 768p or 2K (default). minimax/h3-max: MiniMax H3 Max, a distinct post-trained model with faster generation, stronger prompt adherence, and improved aesthetics. Supports text, first/last frames, and image/video/audio references. tier=turbo selects the faster, half-price text/image arm; reference inputs use standard. Duration: 5-15s. Resolution: 480p or 768p. wan/v2.7: Wan v2.7 with enhanced motion smoothness and scene fidelity. Supports text-to-video, image-to-video, reference-to-video, and edit-video. Ratios: 16:9, 9:16, 1:1. Duration: 5s. Resolution: 480p, 720p vidu/q3: Vidu Q3 model with enhanced quality and audio generation. Supports text-to-video, image-to-video, and reference-to-video (reference_mode with 1-4 image_urls keeps subjects/scenes consistent; no turbo tier in reference mode). tier=turbo runs the faster turbo arm at half the price. Supported ratios: 16:9, 9:16, 4:3, 3:4, 1:1. Duration: 1-16s. Resolution: 720p, 1080prunway/gen4_turbo: A model for generate video with high quality, fast.Supported ratios: 5:3, 3:5. Only support i2v. Duration: 5s, 10spixverse/v6: PixVerse V6 latest model with lifelike motion, richer skin detail, real emotions. Full cinematic control including choreography and camera. Supports text-to-video, image-to-video, transition, and video extend generation. VFX, time-lapse, transformation scenes, product demos, 360° views, multi-shot storytelling. Extend: pass video_url to extend an existing video. Supported ratios: 16:9, 9:16, 4:3, 1:1, 3:4. Duration: 5s, 8s. pixverse/c1: PixVerse C1 with native audio (audio_enable=true), text-to-video, image-to-video, transitions (2 images = first/last frame) and reference-to-video (reference_mode with 1-7 image_urls; address them in the prompt as @image1..@image7). No video extend. Supported ratios: 16:9, 9:16, 4:3, 3:4, 1:1, 3:2, 2:3, 21:9. Duration: 1-15s. Resolution: 360p, 540p, 720p, 1080p. fal-ai/bytedance/seedance-2.0: Bytedance Seedance 2.0 model for highest quality video generation with native audio and lip-sync. Supports text-to-video and image-to-video with first/last frame control. Supported ratios: 21:9, 16:9, 4:3, 1:1, 3:4, 9:16. Duration: 4-15s. Resolution: 480p, 720p, or 1080p on the standard tier (~2.25x cost; fast/mini cap at 720p). tier picks the variant: standard, fast, or mini (cheapest, ~half the standard price). Also supports reference-to-video with up to 9 images, 3 videos, or 3 audio refs via settings.fal-ai/bytedance-upscaler/upscale/video: ByteDance Video Upscaler for enhancing video quality. Upscales videos to higher resolutions (2k) Requires video_url parameter. Does NOT generate new videos. Use when user wants to improve quality of existing videos.xai/grok-imagine-video: xAI Grok Imagine Video v1.5 for high-quality video generation. Supports text-to-video and image-to-video generation. 2+ image_urls (or reference_mode=true) = style/character references, up to 7 (reference mode caps at 720p). video_url (MP4, 2-15s source) = video EXTENSION mode: continues from the last frame, 2-10s extension, 720p. 480p/720p/1080p output. Supported ratios: 16:9, 4:3, 3:2, 1:1, 2:3, 3:4, 9:16. Duration: 1-15s. alibaba/happy-horse: Alibaba Happy Horse video model. High-quality generation from text, or image-to-video from a single first-frame image (one image_url without reference_mode). Supported ratios: 16:9, 9:16, 1:1, 4:3, 3:4, 21:9, 9:21, 5:4, 4:5. Duration: 3-15s. Resolution: 720p, 1080p.alibaba/happy-horse/reference-to-video: Alibaba Happy Horse reference-to-video. Generate video from 1-9 reference images; address subjects in the prompt as character1..character9 (order matches image_urls). Requires image_urls. Supported ratios: 16:9, 9:16, 1:1, 4:3, 3:4, 21:9, 9:21, 5:4, 4:5. Duration: 3-15s. Resolution: 720p, 1080p.alibaba/happy-horse/video-edit: Alibaba Happy Horse video-edit. Edit a source video using a text prompt; optional 1-5 reference images addressed as @Image1..@Image5. Requires video_url (MP4/MOV, 3-60s). Output preserves source aspect ratio, capped at 15s. Resolution: 720p, 1080p.gemini/omni-flash: Google's multimodal video model with native audio. Supports text-to-video, image-to-video (1 image as first frame), reference-to-video (set reference_mode with image_urls to use images as subject/style references — even a single image; 2+ images are references by default), and video editing (pass video_url to edit an existing clip — restyle, restage, add objects). Bind image roles in the prompt with <FIRST_FRAME> / <IMAGE_REF_0>. Ratios: 16:9, 9:16. Duration: 3-10s. Resolution: 720p. bfl/flux-3-preview-high: BFL FLUX 3 video model with native audio. Supports text-to-video and 1-10 ordered keyframes, or continuation from one source video selected with input_mode=start_video. Ratios: auto, 21:9, 2:1, 16:9, 4:3, 1:1, 3:4, 9:16. Duration: 5-20s. Resolution: 720p, 1080p. Only one input family may be used per request.fal-ai/bytedance/seedance-2.5: Bytedance Seedance 2.5 for native 30-second single-shot video with synchronized audio. Supports text-to-video and image-to-video with first/last frame control. No tier variants. Supported ratios: auto, 21:9, 16:9, 4:3, 1:1, 3:4, 9:16. Duration: 4-30s, or 0 = auto (needs video refs). Resolution: 480p, 720p, or 1080p (~2.46x cost). Also supports reference-to-video (up to 30 images, 10 videos, 10 audio refs; no image required), video EDITING and EXTENSION — source in video_urls, the prompt describes the change or continuation.wan/v3.0: Alibaba Wan 3.0 — next-gen open-weight Wan with native audio and single-pass clips up to 30s. Supports text-to-video, image-to-video with first/last frame control, and reference-to-video (reference_mode with up to 10 reference images, 5 videos and 5 audio clips, each ref modality <=15s combined — address them positionally in the prompt as 'Image 1', 'Video 1', 'Audio 1'). tier=fast runs the accelerated Prime arm (~1.4x cost). Supported ratios: adaptive, 16:9, 4:3, 1:1, 3:4, 9:16. Duration: 2-30s. Resolution: 480p, 720p, 1080p. Duration ranges above are hard caps; out-of-range values are clamped to the nearest supported value. (string, one of: kling/v3, kling/o3, gemini/veo3.1, minimax/h3, minimax/h3-max, wan/v2.7, vidu/q3, runway/gen4_turbo, pixverse/v6, pixverse/c1, fal-ai/bytedance/seedance-2.0, fal-ai/bytedance-upscaler/upscale/video, xai/grok-imagine-video, alibaba/happy-horse, alibaba/happy-horse/reference-to-video, alibaba/happy-horse/video-edit, gemini/omni-flash, bfl/flux-3-preview-high, fal-ai/bytedance/seedance-2.5, wan/v3.0) |
|
|
@@ -47,6 +48,8 @@ gsk video [options]
|
|
|
47
48
|
| `--audio_enable` | No | Set true when the user wants audio, voiceover, or narration. kling/v3 outputs a SILENT video unless true (audio costs more per second). Omit to keep each model's default (kling: off; seedance/veo: on). (boolean) |
|
|
48
49
|
| `--tier` | No | Model tier (options are model-specific — check get_model_info tier_options). Seedance 2.0: standard \| fast (quicker/cheaper, 720p cap) \| mini (cheapest, ~half the standard price, 720p cap). Veo 3 / Veo 3.1 family: standard \| fast (quicker, lower cost). Kling V3: pro \| standard \| turbo (fastest, no audio/frame control). Kling O3/Motion Control: pro \| standard. Vidu Q3: standard \| turbo (half price). Wan 3.0: standard \| fast (the accelerated Prime arm — quicker but ~1.4x the price). MiniMax H3 Max: standard \| turbo (faster, half price; references use standard). MiniMax H3 ignores tier. Models without tiers ignore it. (string, one of: standard, fast, mini, pro, turbo, max) |
|
|
49
50
|
| `--video_size` | No | Output resolution. 'auto' uses the model default (usually 720p). MiniMax H3 defaults to 2K with a cheaper 768p tier; MiniMax H3 Max uses 480p or 768p. Values a model does not publish fall back to its default. Higher tiers bill more per second — see each model's Resolution line. Default: auto. (string, one of: auto, 480p, 720p, 768p, 1080p, 2K, 4k, default: `auto`) |
|
|
51
|
+
| `--draft` | No | Seedance 2.5 only: render a 480p draft (billed at the 480p rate) to check structure, motion and prompt fit before the full render. Ignored on other models. (boolean, default: `False`) |
|
|
52
|
+
| `--draft_task_id` | No | task_id of a successful Seedance 2.5 draft (within 7 days): renders its final 1080p video, reusing the draft's prompt, references, duration and ratio. Send no other inputs and no draft flag with it; bills the 1080p rate. (string) |
|
|
50
53
|
| `--reference_mode` | No | Enable reference-to-video mode: provided images guide character/style rather than being used as the first frame. Requires image_urls. For Seedance 2.0: routes to seedance-2.0/ref (combinable with the fast/mini tiers, supports up to 9 reference images plus optional reference video/audio). For Veo 3.1 base model: upgrades to veo3.1/reference-to-video. For MiniMax H3: treats images as subject/style references instead of keyframes (any video_urls/audio_urls also force reference mode). For Vidu Q3: routes to the reference-to-video mix endpoint (1-4 reference images). For PixVerse C1: 1-7 subject references (@image1..@image7 in the prompt). For Grok Imagine Video: 1-7 references (2+ images switch automatically; caps output at 720p). For Kling O3: 1-4 references incl. elements (3+ images switch automatically). Ignored on other models. Default: false. (boolean, default: `False`) |
|
|
51
54
|
|
|
52
55
|
## Local File Support
|
|
@@ -1,10 +1,11 @@
|
|
|
1
1
|
---
|
|
2
2
|
name: gsk-voice-cloning
|
|
3
3
|
version: 1.0.0
|
|
4
|
-
description: Clone a voice from an audio sample
|
|
5
|
-
for
|
|
4
|
+
description: Clone a voice from an audio sample (MiniMax or Gemini voice replication).
|
|
5
|
+
Returns a voice ID for a follow-up TTS call.
|
|
6
6
|
metadata:
|
|
7
7
|
category: general
|
|
8
|
+
generated_by: genspark_tool_cli/generate_skills.py
|
|
8
9
|
requires:
|
|
9
10
|
bins:
|
|
10
11
|
- gsk
|
|
@@ -15,7 +16,7 @@ metadata:
|
|
|
15
16
|
|
|
16
17
|
**PREREQUISITE:** Read `../gsk-shared/SKILL.md` for auth, global flags, and security rules.
|
|
17
18
|
|
|
18
|
-
Clone a voice from an audio sample
|
|
19
|
+
Clone a voice from an audio sample (MiniMax or Gemini voice replication). Returns a voice ID for a follow-up TTS call.
|
|
19
20
|
|
|
20
21
|
## Usage
|
|
21
22
|
|
|
@@ -29,7 +30,7 @@ gsk voice-clone [options]
|
|
|
29
30
|
|
|
30
31
|
| Flag | Required | Description |
|
|
31
32
|
|------|----------|-------------|
|
|
32
|
-
| `<query>` (positional) | No | A
|
|
33
|
+
| `<query>` (positional) | No | A name for the cloned voice. (string) |
|
|
33
34
|
| `-m`, `--model` | Yes | fal-ai/minimax/voice-clone: Clone a voice from a sample audio using Minimax. Requires a single audio sample (min 10 seconds). Returns a custom_voice_id for use with fal-ai/minimax/speech-2.8-hd. (string, one of: fal-ai/minimax/voice-clone) |
|
|
34
35
|
| `-f`, `--voice_files` | Yes | URL or path to voice sample audio file (minimum 10 seconds). (array) |
|
|
35
36
|
|
|
@@ -6,6 +6,7 @@ description: 'YouTube operations. Actions: search (find videos by keyword), tran
|
|
|
6
6
|
billed 1 credit/MB capped at 1024 credits, 1 GB cap; use -o to save locally).'
|
|
7
7
|
metadata:
|
|
8
8
|
category: general
|
|
9
|
+
generated_by: genspark_tool_cli/generate_skills.py
|
|
9
10
|
requires:
|
|
10
11
|
bins:
|
|
11
12
|
- gsk
|