ghosty-acp 0.0.68 → 0.0.70

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "ghosty-acp",
3
- "version": "0.0.68",
3
+ "version": "0.0.70",
4
4
  "description": "Conecta tu editor a un agente ACP remoto. Puente entre entrada/salida estándar y WebSocket, sin dependencias.",
5
5
  "type": "module",
6
6
  "bin": {
@@ -5,7 +5,7 @@ license: MIT
5
5
  compatibility: Needs Node 22+ (for npx @ghostystudio/cli) or curl, and network access to https://www.ghosty.studio
6
6
  metadata:
7
7
  author: ghosty-studio
8
- version: "1.99"
8
+ version: "1.101"
9
9
  ---
10
10
 
11
11
  # Configure a Ghosty Studio agent
@@ -216,7 +216,7 @@ legacy and always `true`.)
216
216
  | "who changed this?", "audit log", "what happened to this agent's channels?", "who logged in as whom" | `ghosty audit ls [--workspace slug] [--agent id] [--actor email] [--action agent\|channel\|card\|factory\|workspace] [--since 24h] --json` (staff see all; others only their workspaces/agents) |
217
217
  | "why did this run take so long?", "what did run N cost?", "where did @build spend its time?" | `ghosty factory runs show <n> --timing --workspace <slug> --json` (each role's turns with minutes, model, USD; clock split working / waiting for a person / other; commands ≥ 60 s). Long commands without output usually mean a hung test or a missing service in the work box |
218
218
  | "why did run N fail / get stuck?", "what should we change so it doesn't happen again?" | `ghosty factory runs why <n> --workspace <slug> --json` (category — `approved_unmet_criterion` first when a plan criterion like "in the preview" was approved without evidence, see `acceptance[]` —, impact, critical step with evidence, typed actions: role_rules → `.ghosty/factory.md`, repo_setup, skill, prompt; `--refresh` redoes it). Stuck patterns per step are in `runs show <n> --timing` |
219
- | "the run is escalated, what do I pick?", "a or b", "merge it as it is" | `ghosty factory runs choose <n> --workspace <slug>` lists the 2–3 paths @check left (a/b/c with their action); `ghosty factory runs choose <n> <letter> --yes` applies one and the choice reaches the next role word for word. `merge_as_is` merges with red CI that the PR did not cause (logged as `merge_override`): confirm with the user first |
219
+ | "the run is escalated, what do I pick?", "a or b", "merge it as it is" | `ghosty factory runs choose <n> --workspace <slug>` lists the 2–3 paths @check left (a/b/c with their action); `ghosty factory runs choose <n> <letter> --yes` applies one and the choice reaches the next role word for word. `merge_as_is` merges with red CI that the PR did not cause (logged as `merge_override`): confirm with the user first. `fix_main` opens a separate run that fixes the main branch (it waits for the owner's signature); the escalated run then shows «Waiting for the main fix (#N)» and resumes on its own when that fix merges |
220
220
  | "how is the factory doing?", "cost per merged PR", "merge rate" | `ghosty factory runs stats --since 30d [--repo owner/repo] --workspace <slug> --json` |
221
221
  | "connect this repo to the factory", "create a factory room", "set up the factory for <repo>" | `ghosty factory rooms connect owner/repo --room <name> --create --workspace <slug> --json` (checks the GitHub App sees it; returns `readiness` = what's missing); `ghosty factory rooms ls --workspace <slug>`; `rooms disconnect … --yes` |
222
222
  | "hand this to the factory", "let the factory build it", "delegate this PR", "send it to @plan" | `ghosty factory run "<goal, why, how we know it's done>" --repo owner/repo --wait --workspace <slug> --json` → it returns at `plan_review` with `next` (the commands). Show the user the plan (`runs show <n> --full`) and ask before `runs approve <n> --plan-version N --yes`; then `runs watch <n>` until `pr_review` and ask before `runs merge <n> --yes`. `runs changes <n> --note "…"` sends it back; `runs card <n>` brings the live card down. More in the `ghosty-factory` skill |
@@ -231,6 +231,7 @@ legacy and always `true`.)
231
231
  | "have it speak in this Teams room / thread at 9" | `ghosty schedule add <id> "…" --teams <workspace>#<room>[/<thread-message-id>] --at ISO \| --in 30m` → it answers there as if mentioned; `ghosty schedule ls\|rm <id> [sched-id] --teams <workspace>` |
232
232
  | "have it do X every day / at 9 / remind me" | `ghosty schedule add <id> "…" --at ISO \| --in 30m [--every 12h --until ISO] [--title T]`; each run notifies the user's phone, an answer of exactly `OK` stays silent. `ghosty schedule ls|rm <id>` |
233
233
  | "watch this page / tell me when it changes / track the competitor's prices" | `ghosty watch add <id> <url> [--every 24h] [--focus "heading or line"] [--instruction "what matters"] [--title T]` — the platform fetches it without a model turn and only opens a turn when it changed (never a `schedule --every` that checks a page: each tick is a paid turn). `ghosty watch check <id> <watch-id> --dry-run` shows the diff now without saving; `watch ls\|rm\|resume <id> [watch-id]`. Inside a chat the agent has `watch_add`/`watch_list`/`watch_remove` |
234
+ | "my morning brief", "stop the morning summary", "send it at 8", "without the audio" | `ghosty brief` (on/off, voice, time, days, last sent); `ghosty brief on\|off`; `ghosty brief --at 8:15 --days 1-5 --voice off` (any subset; `--days all\|weekdays\|0,6`). `ghosty brief run --dry-run` builds today's with its voice note and prints it (sends nothing); `brief run` sends it now. It is the ONLY morning notice: lands in the latest personal chat with a voice note and a push, only when there's something. Inside a chat the agent has `daily_brief_set` |
234
235
  | "how does the voice sound / say this in the voice notes' voice / does it have a Mexican accent" | `ghosty tts "text" [--voice em_santa\|em_alex\|ef_dora] [--out voz.wav]` then `afplay voz.wav` — the house voice (Kokoro, Mexican Spanish es-419, pronunciation lexicon: write brand names normally, never phonetically). Up to 400 characters |
235
236
 
236
237
  Read `references/api.md` for exact request/response shapes before calling.