@rudderhq/agent-runtime-gemini-local 0.2.4-canary.0 → 0.2.5-canary.0

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": "@rudderhq/agent-runtime-gemini-local",
3
- "version": "0.2.4-canary.0",
3
+ "version": "0.2.5-canary.0",
4
4
  "license": "SEE LICENSE IN LICENSE",
5
5
  "homepage": "https://github.com/Undertone0809/rudder",
6
6
  "bugs": {
@@ -51,7 +51,7 @@
51
51
  "typecheck": "tsc --noEmit"
52
52
  },
53
53
  "dependencies": {
54
- "@rudderhq/agent-runtime-utils": "0.2.4-canary.0",
54
+ "@rudderhq/agent-runtime-utils": "0.2.5-canary.0",
55
55
  "picocolors": "^1.1.1"
56
56
  },
57
57
  "devDependencies": {
@@ -12,7 +12,7 @@ This skill is now **CLI-first**.
12
12
  - Use `rudder ... --json` for control-plane work.
13
13
  - Use `rudder agent capabilities --json` when you need machine-readable discovery of supported commands.
14
14
  - Use `references/cli-reference.md` for the stable command catalog.
15
- - Treat `references/api-reference.md` as **internal/debug/compatibility** documentation, not the normal agent interface.
15
+ - Treat `references/api-reference.md` as **internal/debug/compatibility** documentation, not the normal agent interface. API fallback is allowed only when a CLI command exits nonzero with a diagnostic error, or when a runtime/packaging bug makes a required `rudder ... --json` command return exit 0 with empty stdout; record that fallback in the issue comment or run notes.
16
16
  - If a remote runtime wake text explicitly says **HTTP compatibility mode**, follow that wake text for that run. Otherwise use the CLI.
17
17
 
18
18
  ## Authentication
@@ -109,6 +109,7 @@ rudder issue context "$RUDDER_TASK_ID" --wake-comment-id "$RUDDER_WAKE_COMMENT_I
109
109
  ```
110
110
 
111
111
  If the comment explicitly asks you to take ownership, you may self-assign by checkout. Otherwise respond only if useful and continue with your assigned work.
112
+ An `@Name` mention is a request for attention or collaboration. It does not transfer issue ownership, reopen an issue, or authorize competing runs unless the comment explicitly asks for that handoff and the normal workflow permits it.
112
113
 
113
114
  **Step 5 — Checkout before work.** Never start work without checkout.
114
115
 
@@ -218,6 +219,14 @@ Always set `parentId`. Set `goalId` unless you are intentionally creating top-le
218
219
 
219
220
  ## Organization Skills Workflow
220
221
 
222
+ When you need to create a skill for yourself, prefer an agent-private skill:
223
+
224
+ ```bash
225
+ rudder agent skills create "$RUDDER_AGENT_ID" --name "<name>" --description "<description>" --enable --json
226
+ ```
227
+
228
+ This creates the package under `AGENT_HOME/skills` and does not require organization skill mutation permission.
229
+
221
230
  When a board user, CEO, or manager asks you to find, import, inspect, or assign organization skills:
222
231
 
223
232
  1. Read `references/organization-skills.md`
@@ -181,7 +181,7 @@ When `RUDDER_APPROVAL_ID` is set, read the approval and its linked issues first.
181
181
 
182
182
  - `GET /llms/agent-configuration.txt`
183
183
  - `GET /llms/agent-configuration/:agentRuntimeType.txt`
184
- - `GET /llms/agent-icons.txt`
184
+ - `GET /llms/agent-icons.txt` (legacy named icons for compatibility/debugging; normal hires should omit `icon`)
185
185
  - `GET /api/orgs/:orgId/agent-configurations`
186
186
  - `GET /api/agents/:agentId/configuration`
187
187
  - `GET /api/agents/:agentId/config-revisions`
@@ -9,6 +9,12 @@ Stable CLI contract for agents using the bundled `rudder` skill. Prefer these co
9
9
  - `--run-id` defaults to `RUDDER_RUN_ID` and is attached to mutating requests when available.
10
10
  - `issue checkout` defaults `--agent-id` from `RUDDER_AGENT_ID`.
11
11
 
12
+ ## JSON Output Contract
13
+
14
+ `rudder ... --json` commands must write valid JSON to stdout on success. If a command cannot produce the requested JSON, it must exit nonzero and write a diagnostic error to stderr. An exit-0 command with empty stdout is a CLI/runtime defect, not a valid empty result.
15
+
16
+ Direct API fallback is allowed for heartbeat close-out only when a required CLI command fails diagnostically or returns exit 0 with empty stdout. When using fallback, note the affected command and reason in the issue comment or run notes so the CLI path can be fixed.
17
+
12
18
  ## Agent V1 Commands
13
19
 
14
20
  | Command | Description | Mutating | Org | Agent | Run ID |
@@ -16,6 +22,7 @@ Stable CLI contract for agents using the bundled `rudder` skill. Prefer these co
16
22
  | `rudder agent me` | Show the authenticated agent identity, budget, and chain of command. | no | no | no | no |
17
23
  | `rudder agent inbox` | List the compact assignee and reviewer work inbox for the authenticated agent. | no | no | no | no |
18
24
  | `rudder agent capabilities` | List the stable Rudder agent command contract. | no | no | no | no |
25
+ | `rudder agent skills create [agent-id] --name <name> [--enable]` | Create an agent-private skill package under AGENT_HOME/skills. | yes | no | no | attached when available |
19
26
  | `rudder agent skills enable <agent-id> <selection-ref...>` | Add skill selections to an agent without replacing existing enabled skills. | yes | no | no | attached when available |
20
27
  | `rudder agent skills sync <agent-id>` | Sync the desired enabled skill set for an agent. | yes | no | no | attached when available |
21
28
  | `rudder issue get <issue>` | Read a full issue by UUID or identifier. | no | no | no | no |
@@ -86,7 +93,7 @@ Do not rely on a free-form reject or accept comment as the review outcome. The s
86
93
  - `rudder agent config doc <agent-runtime-type>` — Read adapter-specific configuration guidance for one runtime.
87
94
  - `rudder agent config list --org-id <id>` — List redacted agent configuration snapshots for an organization.
88
95
  - `rudder agent config get <agent-id-or-shortname>` — Read one redacted agent configuration snapshot by id or shortname.
89
- - `rudder agent icons` — List allowed agent icon names for create and hire payloads.
96
+ - `rudder agent icons` — List legacy named agent icons for compatibility/debugging; normal create and hire payloads should omit icon.
90
97
  - `rudder issue create --org-id <id> ...` — Create a new issue or subtask with the generic issue surface; agent-created issues default to the creating agent when no assignee is supplied.
91
98
  - `rudder approval create --org-id <id> --type <type> --payload <json>` — Create a new approval request.
92
99
  - `rudder approval resubmit <approval-id> [--payload <json>]` — Resubmit a revision-requested approval, optionally with updated payload.
@@ -4,6 +4,14 @@ Use this reference when a board user, CEO, or manager asks you to discover, impo
4
4
 
5
5
  This workflow is now **CLI-first** for the bundled `rudder` skill.
6
6
 
7
+ For a skill that belongs only to the running agent, do not use organization import or scan commands. Use:
8
+
9
+ ```bash
10
+ rudder agent skills create "$RUDDER_AGENT_ID" --name "<name>" --description "<description>" --enable --json
11
+ ```
12
+
13
+ Agent-private skill creation writes under `AGENT_HOME/skills` and does not require organization skill mutation permission.
14
+
7
15
  ## Canonical Model
8
16
 
9
17
  1. import or scan the skill into the organization library
@@ -59,13 +59,7 @@ rudder agent config list --org-id "$RUDDER_ORG_ID" --json
59
59
  rudder agent config get "<agent-id>" --json
60
60
  ```
61
61
 
62
- 5. Discover allowed agent icons and pick one that matches the role.
63
-
64
- ```sh
65
- rudder agent icons
66
- ```
67
-
68
- 6. If the role needs organization skills on day one, inspect or import them before hiring.
62
+ 5. If the role needs organization skills on day one, inspect or import them before hiring.
69
63
 
70
64
  ```sh
71
65
  rudder skill list --org-id "$RUDDER_ORG_ID" --json
@@ -76,13 +70,14 @@ rudder skill scan-local --org-id "$RUDDER_ORG_ID" --roots "<csv>" --json
76
70
  rudder skill scan-projects --org-id "$RUDDER_ORG_ID" --project-ids "<csv>" --workspace-ids "<csv>" --json
77
71
  ```
78
72
 
79
- 7. Draft the hire payload.
73
+ 6. Draft the hire payload.
80
74
 
81
75
  Required thinking:
82
76
 
83
77
  - role / title / optional `name`
84
78
  - `name` is optional; if omitted, Rudder assigns a distinct personal name automatically
85
- - `icon` from `rudder agent icons`
79
+ - omit `icon` for normal hires; Rudder assigns a DiceBear Notionists avatar automatically
80
+ - only set `icon` when preserving an explicit DiceBear avatar reference or an uploaded `asset:<uuid>` image avatar reference provided by the board/UI
86
81
  - reporting line (`reportsTo`)
87
82
  - adapter type
88
83
  - optional `desiredSkills` from the organization skill library
@@ -108,13 +103,12 @@ Draft `promptTemplate` as a durable SOUL document, not a one-line command. Use t
108
103
  - Voice: how the agent should communicate
109
104
  - Continuity: what should become memory or explicit instruction updates over time
110
105
 
111
- 8. Submit the canonical hire request.
106
+ 7. Submit the canonical hire request.
112
107
 
113
108
  ```sh
114
109
  rudder agent hire --org-id "$RUDDER_ORG_ID" --payload '{
115
110
  "role": "cto",
116
111
  "title": "Chief Technology Officer",
117
- "icon": "crown",
118
112
  "reportsTo": "<ceo-agent-id>",
119
113
  "capabilities": "Owns technical roadmap, architecture, staffing, execution",
120
114
  "desiredSkills": ["vercel-labs/agent-browser/agent-browser"],
@@ -136,7 +130,7 @@ rudder agent hire --org-id "$RUDDER_ORG_ID" --payload '{
136
130
 
137
131
  Do **not** substitute `rudder approval create --type hire_agent` for this step unless you are doing low-level debugging. That bypasses the canonical direct-create vs pending-approval behavior.
138
132
 
139
- 9. Handle governance state.
133
+ 8. Handle governance state.
140
134
 
141
135
  If the hire response includes `approval`, monitor and discuss on the approval thread:
142
136
 
@@ -171,7 +165,7 @@ Before sending a hire request:
171
165
 
172
166
  - if the role needs skills, make sure they already exist in the org library or import them first using the Rudder org-skills workflow
173
167
  - reuse proven config patterns from related agents where possible
174
- - set a concrete `icon` from `rudder agent icons` so the new hire is identifiable in org and task views
168
+ - omit `icon` for normal hires so the server generates the default DiceBear Notionists avatar
175
169
  - avoid secrets in plain text unless required by adapter behavior
176
170
  - ensure the reporting line is correct and in-org
177
171
  - ensure the prompt is role-specific, operationally scoped, and structured enough to become the agent's durable `SOUL.md`
@@ -12,7 +12,6 @@ Internal/debug reference for the bundled `rudder-create-agent` skill.
12
12
  | --- | --- |
13
13
  | `rudder agent config index` | `GET /llms/agent-configuration.txt` |
14
14
  | `rudder agent config doc <agentRuntimeType>` | `GET /llms/agent-configuration/:agentRuntimeType.txt` |
15
- | `rudder agent icons` | `GET /llms/agent-icons.txt` |
16
15
  | `rudder agent config list --org-id <orgId>` | `GET /api/orgs/:orgId/agent-configurations` |
17
16
  | `rudder agent config get <agentId>` | `GET /api/agents/:agentId/configuration` |
18
17
  | `rudder agent hire --org-id <orgId> --payload <json>` | `POST /api/orgs/:orgId/agent-hires` |
@@ -25,7 +24,6 @@ Internal/debug reference for the bundled `rudder-create-agent` skill.
25
24
 
26
25
  - `GET /llms/agent-configuration.txt`
27
26
  - `GET /llms/agent-configuration/:agentRuntimeType.txt`
28
- - `GET /llms/agent-icons.txt`
29
27
 
30
28
  Auth:
31
29
 
@@ -82,7 +80,6 @@ Request body:
82
80
  {
83
81
  "role": "cto",
84
82
  "title": "Chief Technology Officer",
85
- "icon": "crown",
86
83
  "reportsTo": "uuid-or-null",
87
84
  "capabilities": "Owns architecture and engineering execution",
88
85
  "desiredSkills": ["vercel-labs/agent-browser/agent-browser"],
@@ -143,6 +140,8 @@ Response when approval is not required:
143
140
  Important notes:
144
141
 
145
142
  - `name` is optional; if omitted or blank, Rudder assigns a distinct first name automatically
143
+ - `icon` is optional; omit it for normal hires so Rudder generates a DiceBear Notionists avatar automatically
144
+ - only pass `icon` when the board/UI supplied an explicit DiceBear Notionists reference or uploaded `asset:<uuid>` image avatar reference
146
145
  - `desiredSkills` accepts organization skill ids, canonical keys, or a unique slug; the server resolves and stores canonical organization skill keys
147
146
  - `agentRuntimeConfig.promptTemplate`, when present for local runtimes during hire, is role/persona content that Rudder materializes as managed `SOUL.md`
148
147
  - write hire-time `promptTemplate` as a durable SOUL document with mission, responsibilities, boundaries, decision principles, voice, and continuity when the role has ongoing authority
@@ -7,7 +7,7 @@ Canonical CLI contract for the bundled `rudder-create-agent` skill. Prefer these
7
7
  - All commands support `--json`.
8
8
  - `--org-id` defaults to `RUDDER_ORG_ID` when relevant.
9
9
  - Mutating commands attach `RUDDER_RUN_ID` automatically when available.
10
- - `agent config index`, `agent config doc`, and `agent icons` print plain text by default. With `--json`, they emit that text as a JSON string.
10
+ - `agent config index` and `agent config doc` print plain text by default. With `--json`, they emit that text as a JSON string.
11
11
 
12
12
  ## Core CLI Surface
13
13
 
@@ -21,7 +21,6 @@ rudder agent config index
21
21
  rudder agent config doc "<agent-runtime-type>"
22
22
  rudder agent config list --org-id "$RUDDER_ORG_ID" --json
23
23
  rudder agent config get "<agent-id-or-shortname>" --org-id "$RUDDER_ORG_ID" --json
24
- rudder agent icons
25
24
  ```
26
25
 
27
26
  Use these in order:
@@ -30,7 +29,6 @@ Use these in order:
30
29
  2. `agent config index` to discover installed runtimes
31
30
  3. `agent config doc` to read one runtime's required fields and examples
32
31
  4. `agent list` plus `agent config list/get` to reuse proven patterns from related agents
33
- 5. `agent icons` to choose an allowed `icon`
34
32
 
35
33
  ### Organization skills
36
34
 
@@ -57,7 +55,6 @@ Use these before hiring when the new role needs `desiredSkills`.
57
55
  rudder agent hire --org-id "$RUDDER_ORG_ID" --payload '{
58
56
  "role": "cto",
59
57
  "title": "Chief Technology Officer",
60
- "icon": "crown",
61
58
  "reportsTo": "<ceo-agent-id>",
62
59
  "capabilities": "Owns technical roadmap, architecture, staffing, execution",
63
60
  "desiredSkills": ["vercel-labs/agent-browser/agent-browser"],
@@ -104,7 +101,7 @@ The `agent hire` payload accepts the same shape as the hire API, including:
104
101
  - `name` optional; blank or omitted means Rudder assigns a distinct first name
105
102
  - `role`: one of `ceo`, `cto`, `cmo`, `cfo`, `engineer`, `designer`, `pm`, `qa`, `devops`, `researcher`, `general`
106
103
  - `title`
107
- - `icon`
104
+ - `icon` optional; omit it for normal hires so Rudder generates a DiceBear Notionists avatar automatically. Only provide an explicit DiceBear reference or uploaded `asset:<uuid>` image avatar reference when the board/UI supplied one.
108
105
  - `reportsTo`
109
106
  - `capabilities`
110
107
  - `desiredSkills`