@zibby/skills 0.1.95 → 0.1.96

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/dist/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@zibby/skills",
3
- "version": "0.1.95",
3
+ "version": "0.1.96",
4
4
  "description": "Built-in skill definitions for the Zibby agent-workflow framework",
5
5
  "type": "module",
6
6
  "main": "dist/index.js",
@@ -20,6 +20,7 @@ Every command lives under `zibby agent <verb>` for consistency. The bare top-lev
20
20
  | [`zibby agent download <uuid>`](#agent-download) | Pull a deployed agent back to local. Edit + redeploy. |
21
21
  | [`zibby agent delete <uuid>`](#agent-delete) | Delete a deployed agent. |
22
22
  | [`zibby agent start <name>`](#agent-start) | Long-lived dev server (Studio integration). Most users want `run`. |
23
+ | [`zibby agent update <uuid>`](#agent-update) | One-shot edit of a deployed agent: model, review @mention, trigger events, name, max runtime. |
23
24
  | [`zibby agent env <verb>`](#agent-env) | Manage per-agent encrypted env vars: `list`, `set`, `unset`, `push`. |
24
25
 
25
26
  Plus the test recipe + memory + project setup:
@@ -149,6 +150,38 @@ Options:
149
150
 
150
151
  **Storage & retention.** Live logs are kept in CloudWatch for 30 days. Beyond that, the per-run session folder (uploaded to S3 at the end of every execution) is the long-term archive — pull it back with `zibby agent download <uuid>`.
151
152
 
153
+ ## agent update {#agent-update}
154
+
155
+ ```bash
156
+ zibby agent update <uuid> --model claude:sonnet-4.6 # switch vendor:model
157
+ zibby agent update <uuid> --mention @zibby # review @mention token
158
+ zibby agent update <uuid> --triggers mention,comment # WHICH events fire a review
159
+ zibby agent update <uuid> --name "FE Review Bot" --max-runtime 30
160
+ ```
161
+
162
+ The ONE consolidated editor for per-agent settings that used to be
163
+ dashboard-only. Flags can be combined; the command echoes the updated
164
+ settings so you can confirm what changed.
165
+
166
+ `--triggers` takes friendly names and controls **when a review agent runs**:
167
+
168
+ | Name | Fires on |
169
+ |---|---|
170
+ | `opened` | a new PR/MR is opened |
171
+ | `commit` | new commits are pushed to an open PR/MR |
172
+ | `mention` | someone @-mentions the agent, or replies in one of its threads |
173
+
174
+ `--triggers mention` is the cost-saving mode — the agent reviews **only when
175
+ asked**. New review agents deploy mention-only by default; existing agents keep
176
+ their current behaviour until you change it.
177
+
178
+ Driving Zibby from an AI editor instead of a shell? The MCP twin is
179
+ **`zibby_update_agent`** with the same fields (`model`, `mentionToken`,
180
+ `triggerEvents`, `displayName`, `maxRuntimeMinutes`).
181
+
182
+ Env vars, custom MCP servers and stores are separate surfaces:
183
+ `zibby agent env …`, `zibby_add_mcp`, `zibby_set_node_stores`.
184
+
152
185
  ## agent env {#agent-env}
153
186
 
154
187
  Per-agent encrypted env vars — KMS-stored on the agent record, injected into the Fargate task at trigger time. Agent env wins over project secrets on conflict.
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@zibby/skills",
3
- "version": "0.1.95",
3
+ "version": "0.1.96",
4
4
  "description": "Built-in skill definitions for the Zibby agent-workflow framework",
5
5
  "type": "module",
6
6
  "main": "dist/index.js",