oh-my-customcode 0.63.1 → 0.64.0
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/dist/cli/index.js
CHANGED
package/dist/index.js
CHANGED
package/package.json
CHANGED
|
@@ -11,15 +11,26 @@ Location: `.claude/agents/{name}.md` (single file, kebab-case)
|
|
|
11
11
|
```yaml
|
|
12
12
|
name: agent-name # Unique identifier (kebab-case)
|
|
13
13
|
description: Brief desc # One-line summary
|
|
14
|
-
model: sonnet # sonnet | opus | haiku (or full ID: claude-sonnet-4-6)
|
|
14
|
+
model: sonnet # sonnet | opus | haiku | opusplan (or full ID: claude-sonnet-4-6, claude-opus-4-6[1m])
|
|
15
15
|
tools: [Read, Write, ...] # Allowed tools
|
|
16
16
|
```
|
|
17
17
|
|
|
18
|
+
### Model Aliases
|
|
19
|
+
|
|
20
|
+
| Alias | Full ID | Use Case |
|
|
21
|
+
|-------|---------|----------|
|
|
22
|
+
| `haiku` | claude-haiku-4-5 | Fast, cheap tasks (search, simple edits) |
|
|
23
|
+
| `sonnet` | claude-sonnet-4-6 | General tasks, code generation (default) |
|
|
24
|
+
| `opus` | claude-opus-4-6 | Complex reasoning, architecture |
|
|
25
|
+
| `opusplan` | claude-opus-4-6 + plan mode | Architecture planning with approval gates |
|
|
26
|
+
|
|
27
|
+
Extended context suffix: `[1m]` (e.g., `claude-opus-4-6[1m]`) — enables 1M token context window.
|
|
28
|
+
|
|
18
29
|
### Optional Frontmatter
|
|
19
30
|
|
|
20
31
|
```yaml
|
|
21
32
|
memory: project # user | project | local
|
|
22
|
-
effort: high # low | medium | high
|
|
33
|
+
effort: high # low | medium | high | default | max
|
|
23
34
|
skills: [skill-1, ...] # Skill name references
|
|
24
35
|
source: # For external agents
|
|
25
36
|
type: external
|
|
@@ -144,6 +155,26 @@ Skills persist output to `.claude/outputs/sessions/{YYYY-MM-DD}/{skill-name}-{HH
|
|
|
144
155
|
|
|
145
156
|
Agent body: purpose, capabilities overview, workflow. NOT detailed instructions or reference docs.
|
|
146
157
|
|
|
158
|
+
## Fast Mode
|
|
159
|
+
|
|
160
|
+
Fast Mode uses the same model with faster output. Activated via `/fast` toggle or `fastMode` setting. Does NOT switch to a different model.
|
|
161
|
+
|
|
162
|
+
| Aspect | Normal | Fast Mode |
|
|
163
|
+
|--------|--------|-----------|
|
|
164
|
+
| Model | As configured | Same model |
|
|
165
|
+
| Output speed | Standard | ~2.5x faster |
|
|
166
|
+
| Reasoning depth | Full | Reduced |
|
|
167
|
+
|
|
168
|
+
### Activation
|
|
169
|
+
|
|
170
|
+
- `/fast` — toggle in current session
|
|
171
|
+
- `fastMode: true` in settings.json
|
|
172
|
+
- `CLAUDE_CODE_DISABLE_FAST_MODE=1` — env var to disable
|
|
173
|
+
|
|
174
|
+
### Interaction with Effort
|
|
175
|
+
|
|
176
|
+
When Fast Mode is active, it reduces effective reasoning depth but does NOT override the `effort` frontmatter field. The effort field controls task complexity allocation; Fast Mode controls output generation speed.
|
|
177
|
+
|
|
147
178
|
## Skill Frontmatter
|
|
148
179
|
|
|
149
180
|
Location: `.claude/skills/{name}/SKILL.md`
|
|
@@ -163,7 +194,17 @@ context: fork # Forked context for isolated execution
|
|
|
163
194
|
version: 1.0.0 # Semantic version
|
|
164
195
|
user-invocable: false # Whether user can invoke directly
|
|
165
196
|
disable-model-invocation: true # Prevent model from auto-invoking
|
|
166
|
-
effort: medium # low | medium | high — overrides model effort level when invoked
|
|
197
|
+
effort: medium # low | medium | high | default | max — overrides model effort level when invoked
|
|
198
|
+
argument-hint: "<arg> [--flag]" # CLI-style usage hint displayed in /help and command listings
|
|
199
|
+
model: sonnet # Override spawned model when skill is invoked via Agent
|
|
200
|
+
agent: mgr-creator # Preferred agent to execute this skill
|
|
201
|
+
hooks: # Skill-specific hooks (same syntax as agent hooks)
|
|
202
|
+
PreToolUse:
|
|
203
|
+
- matcher: "Bash"
|
|
204
|
+
command: "echo hook"
|
|
205
|
+
paths: ["src/**/*.ts"] # Conditional loading — skill auto-injected when matching files are open
|
|
206
|
+
shell: "bash" # Shell for embedded script execution
|
|
207
|
+
allowed-tools: [Read, Write, Bash] # Restrict tools available during skill execution
|
|
167
208
|
```
|
|
168
209
|
|
|
169
210
|
When both an agent and its invoked skill specify `effort`, the skill's value takes precedence (more specific invocation-time setting).
|
|
@@ -6,10 +6,12 @@
|
|
|
6
6
|
|
|
7
7
|
| Tier | Tools | Policy |
|
|
8
8
|
|------|-------|--------|
|
|
9
|
-
| 1: Always | Read, Glob, Grep | Free use |
|
|
10
|
-
| 2: Default | Write, Edit | State changes explicitly, notify before modifying important files |
|
|
11
|
-
| 3:
|
|
12
|
-
| 4:
|
|
9
|
+
| 1: Always | Read, Glob, Grep, ToolSearch | Free use, read-only |
|
|
10
|
+
| 2: Default | Write, Edit, NotebookEdit | State changes explicitly, notify before modifying important files |
|
|
11
|
+
| 3: Context | Agent, Skill, EnterPlanMode, ExitPlanMode, EnterWorktree, ExitWorktree, LSP, TodoWrite, AskUserQuestion | Context-dependent, no user approval needed |
|
|
12
|
+
| 4: Approval | Bash, WebFetch, WebSearch | Request user approval on first use |
|
|
13
|
+
| 5: Conditional | TeamCreate, TeamDelete, SendMessage, TaskCreate, TaskGet, TaskList, TaskUpdate, TaskStop, TaskOutput | Available when Agent Teams enabled |
|
|
14
|
+
| 6: MCP | ListMcpResourcesTool, ReadMcpResourceTool, CronCreate, CronDelete, CronList, RemoteTrigger | MCP/extension tools, available when servers configured |
|
|
13
15
|
|
|
14
16
|
## File Access
|
|
15
17
|
|
package/templates/manifest.json
CHANGED