opencode-token-tracker 1.0.0

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/LICENSE ADDED
@@ -0,0 +1,21 @@
1
+ MIT License
2
+
3
+ Copyright (c) 2026 tongshuanglong
4
+
5
+ Permission is hereby granted, free of charge, to any person obtaining a copy
6
+ of this software and associated documentation files (the "Software"), to deal
7
+ in the Software without restriction, including without limitation the rights
8
+ to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
9
+ copies of the Software, and to permit persons to whom the Software is
10
+ furnished to do so, subject to the following conditions:
11
+
12
+ The above copyright notice and this permission notice shall be included in all
13
+ copies or substantial portions of the Software.
14
+
15
+ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16
+ IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17
+ FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
18
+ AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19
+ LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
20
+ OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21
+ SOFTWARE.
package/README.md ADDED
@@ -0,0 +1,119 @@
1
+ # opencode-token-tracker
2
+
3
+ Real-time token usage and cost tracking plugin for [OpenCode](https://opencode.ai).
4
+
5
+ ![Toast Screenshot](./docs/toast-demo.png)
6
+
7
+ ## Features
8
+
9
+ - **Real-time Toast notifications** - See token usage and cost after each AI response
10
+ - **Session statistics** - Track cumulative usage across your entire session
11
+ - **Cost calculation** - Automatic cost estimation based on model pricing
12
+ - **JSONL logging** - All usage data saved locally for analysis
13
+ - **Multi-model support** - Claude, GPT, DeepSeek, Gemini, and more
14
+
15
+ ## Installation
16
+
17
+ Add to your OpenCode config file (`~/.config/opencode/opencode.json`):
18
+
19
+ ```json
20
+ {
21
+ "$schema": "https://opencode.ai/config.json",
22
+ "plugin": ["opencode-token-tracker"]
23
+ }
24
+ ```
25
+
26
+ Restart OpenCode and the plugin will be automatically installed.
27
+
28
+ ## Usage
29
+
30
+ Once installed, you'll see Toast notifications after each AI response:
31
+
32
+ ```
33
+ 12.5K tokens
34
+ $0.023 | Session: $0.156
35
+ ```
36
+
37
+ When a session becomes idle, you'll see a summary:
38
+
39
+ ```
40
+ Session: 45.2K tokens
41
+ $0.156 | 8 msgs | 5min
42
+ ```
43
+
44
+ ## Log Files
45
+
46
+ Token usage is logged to:
47
+
48
+ ```
49
+ ~/.config/opencode/logs/token-tracker/tokens.jsonl
50
+ ```
51
+
52
+ Each line is a JSON object:
53
+
54
+ ```json
55
+ {
56
+ "type": "tokens",
57
+ "sessionId": "ses_xxx",
58
+ "messageId": "msg_xxx",
59
+ "agent": "build",
60
+ "model": "claude-opus-4.5",
61
+ "provider": "github-copilot",
62
+ "input": 1500,
63
+ "output": 350,
64
+ "reasoning": 0,
65
+ "cacheRead": 5000,
66
+ "cacheWrite": 0,
67
+ "cost": 0.0234,
68
+ "_ts": 1234567890123
69
+ }
70
+ ```
71
+
72
+ ## Supported Models
73
+
74
+ | Provider | Models |
75
+ |----------|--------|
76
+ | Anthropic | Claude Opus 4.5, Sonnet 4/4.5, Haiku 4/4.5 |
77
+ | OpenAI | GPT-5.x, GPT-4.x, o1, o3 |
78
+ | DeepSeek | deepseek-chat, deepseek-reasoner |
79
+ | Google | Gemini 2.x, 3.x |
80
+
81
+ Unknown models use a default pricing estimate.
82
+
83
+ ## Configuration
84
+
85
+ Currently the plugin works out of the box with no configuration needed.
86
+
87
+ Future versions may support:
88
+ - Custom pricing overrides
89
+ - Toast display options
90
+ - Export formats
91
+
92
+ ## Development
93
+
94
+ ```bash
95
+ # Clone the repo
96
+ git clone https://github.com/tongsh6/opencode-token-tracker.git
97
+ cd opencode-token-tracker
98
+
99
+ # Install dependencies
100
+ npm install
101
+
102
+ # Build
103
+ npm run build
104
+
105
+ # Link for local testing
106
+ npm link
107
+ cd ~/.config/opencode
108
+ npm link opencode-token-tracker
109
+ ```
110
+
111
+ ## License
112
+
113
+ MIT © [tongsh6](https://github.com/tongsh6)
114
+
115
+ ## Related
116
+
117
+ - [OpenCode](https://opencode.ai) - The AI coding assistant
118
+ - [OpenCode Plugins](https://opencode.ai/docs/plugins) - Plugin documentation
119
+ - [oh-my-opencode](https://github.com/code-yeongyu/oh-my-opencode) - OpenCode enhancement plugin
@@ -0,0 +1,3 @@
1
+ import type { Plugin } from "@opencode-ai/plugin";
2
+ export declare const TokenTrackerPlugin: Plugin;
3
+ export default TokenTrackerPlugin;
package/dist/index.js ADDED
@@ -0,0 +1,218 @@
1
+ import { appendFileSync, existsSync, mkdirSync } from "fs";
2
+ import { join } from "path";
3
+ import { homedir } from "os";
4
+ const LOG_DIR = join(homedir(), ".config", "opencode", "logs", "token-tracker");
5
+ const LOG_FILE = join(LOG_DIR, "tokens.jsonl");
6
+ // Prices as of 2026-02 (update as needed)
7
+ const PRICING = {
8
+ // Anthropic Claude
9
+ "claude-opus-4.5": { input: 15, output: 75, cacheRead: 1.5, cacheWrite: 18.75 },
10
+ "claude-sonnet-4.5": { input: 3, output: 15, cacheRead: 0.3, cacheWrite: 3.75 },
11
+ "claude-sonnet-4": { input: 3, output: 15, cacheRead: 0.3, cacheWrite: 3.75 },
12
+ "claude-haiku-4.5": { input: 0.8, output: 4, cacheRead: 0.08, cacheWrite: 1 },
13
+ "claude-haiku-4": { input: 0.8, output: 4, cacheRead: 0.08, cacheWrite: 1 },
14
+ // OpenAI GPT
15
+ "gpt-5.2": { input: 2.5, output: 10 },
16
+ "gpt-5.2-codex": { input: 3, output: 12 },
17
+ "gpt-5.1": { input: 2, output: 8 },
18
+ "gpt-5": { input: 5, output: 15 },
19
+ "gpt-4.1": { input: 2, output: 8 },
20
+ "gpt-4.1-mini": { input: 0.4, output: 1.6 },
21
+ "gpt-4.1-nano": { input: 0.1, output: 0.4 },
22
+ "gpt-4o": { input: 2.5, output: 10 },
23
+ "gpt-4o-mini": { input: 0.15, output: 0.6 },
24
+ "o3": { input: 10, output: 40 },
25
+ "o3-mini": { input: 1.1, output: 4.4 },
26
+ "o1": { input: 15, output: 60 },
27
+ "o1-mini": { input: 1.1, output: 4.4 },
28
+ // DeepSeek
29
+ "deepseek-chat": { input: 0.14, output: 0.28, cacheRead: 0.014 },
30
+ "deepseek-reasoner": { input: 0.55, output: 2.19, cacheRead: 0.055 },
31
+ // Google Gemini
32
+ "gemini-3-pro": { input: 1.25, output: 5 },
33
+ "gemini-3-pro-preview": { input: 1.25, output: 5 },
34
+ "gemini-3-flash": { input: 0.1, output: 0.4 },
35
+ "gemini-2.5-pro": { input: 1.25, output: 5 },
36
+ "gemini-2.5-flash": { input: 0.075, output: 0.3 },
37
+ "gemini-2.0-flash": { input: 0.1, output: 0.4 },
38
+ // Fallback for unknown models
39
+ "_default": { input: 1, output: 4 },
40
+ };
41
+ function getModelPricing(model) {
42
+ // Try exact match first
43
+ if (PRICING[model])
44
+ return PRICING[model];
45
+ // Try partial match (e.g., "claude-opus-4.5" matches "claude-opus-4.5-xxx")
46
+ const modelLower = model.toLowerCase();
47
+ for (const [key, pricing] of Object.entries(PRICING)) {
48
+ if (key !== "_default" && modelLower.includes(key.toLowerCase())) {
49
+ return pricing;
50
+ }
51
+ }
52
+ return PRICING["_default"];
53
+ }
54
+ function calculateCost(model, input, output, cacheRead = 0, cacheWrite = 0) {
55
+ const pricing = getModelPricing(model);
56
+ // Billable input = total input - cache read (cached tokens are charged at cache rate)
57
+ const billableInput = Math.max(0, input - cacheRead);
58
+ const inputCost = (billableInput / 1_000_000) * pricing.input;
59
+ const outputCost = (output / 1_000_000) * pricing.output;
60
+ const cacheReadCost = (cacheRead / 1_000_000) * (pricing.cacheRead ?? pricing.input * 0.1);
61
+ const cacheWriteCost = (cacheWrite / 1_000_000) * (pricing.cacheWrite ?? pricing.input * 1.25);
62
+ return inputCost + outputCost + cacheReadCost + cacheWriteCost;
63
+ }
64
+ const sessionStats = new Map();
65
+ function getOrCreateSessionStats(sessionId) {
66
+ if (!sessionStats.has(sessionId)) {
67
+ sessionStats.set(sessionId, {
68
+ totalInput: 0,
69
+ totalOutput: 0,
70
+ totalReasoning: 0,
71
+ totalCacheRead: 0,
72
+ totalCacheWrite: 0,
73
+ totalCost: 0,
74
+ messageCount: 0,
75
+ startTime: Date.now(),
76
+ });
77
+ }
78
+ return sessionStats.get(sessionId);
79
+ }
80
+ function formatCost(cost) {
81
+ if (cost < 0.01)
82
+ return `$${cost.toFixed(4)}`;
83
+ if (cost < 1)
84
+ return `$${cost.toFixed(3)}`;
85
+ return `$${cost.toFixed(2)}`;
86
+ }
87
+ function formatTokens(tokens) {
88
+ if (tokens >= 1_000_000)
89
+ return `${(tokens / 1_000_000).toFixed(1)}M`;
90
+ if (tokens >= 1_000)
91
+ return `${(tokens / 1_000).toFixed(1)}K`;
92
+ return tokens.toString();
93
+ }
94
+ // ============================================================================
95
+ // Deduplication
96
+ // ============================================================================
97
+ const seen = new Set();
98
+ function isDuplicate(key) {
99
+ if (seen.has(key))
100
+ return true;
101
+ seen.add(key);
102
+ // Cleanup old entries to prevent memory leak
103
+ if (seen.size > 10000) {
104
+ const entries = Array.from(seen);
105
+ entries.slice(0, 5000).forEach(k => seen.delete(k));
106
+ }
107
+ return false;
108
+ }
109
+ // ============================================================================
110
+ // Logging
111
+ // ============================================================================
112
+ function ensureLogDir() {
113
+ if (!existsSync(LOG_DIR)) {
114
+ mkdirSync(LOG_DIR, { recursive: true });
115
+ }
116
+ }
117
+ function logJson(data) {
118
+ ensureLogDir();
119
+ const entry = JSON.stringify({ ...data, _ts: Date.now() }) + "\n";
120
+ appendFileSync(LOG_FILE, entry);
121
+ }
122
+ export const TokenTrackerPlugin = async ({ directory, client }) => {
123
+ logJson({ type: "init", directory });
124
+ return {
125
+ event: async ({ event }) => {
126
+ try {
127
+ // Handle message updates (token tracking)
128
+ if (event.type === "message.updated") {
129
+ const props = event.properties;
130
+ const info = props?.info;
131
+ if (!info?.tokens)
132
+ return;
133
+ const messageId = info.id;
134
+ const sessionId = info.sessionID;
135
+ if (!messageId || !sessionId)
136
+ return;
137
+ const input = info.tokens.input ?? 0;
138
+ const output = info.tokens.output ?? 0;
139
+ const reasoning = info.tokens.reasoning ?? 0;
140
+ const cacheRead = info.tokens.cache?.read ?? 0;
141
+ const cacheWrite = info.tokens.cache?.write ?? 0;
142
+ const hasTokens = input > 0 || output > 0;
143
+ if (!hasTokens)
144
+ return;
145
+ const dedupeKey = `${messageId}-${input}-${output}`;
146
+ if (isDuplicate(dedupeKey))
147
+ return;
148
+ const model = info.model?.modelID ?? info.modelID ?? "unknown";
149
+ const provider = info.model?.providerID ?? info.providerID ?? "unknown";
150
+ const cost = calculateCost(model, input, output, cacheRead, cacheWrite);
151
+ // Update session stats
152
+ const stats = getOrCreateSessionStats(sessionId);
153
+ stats.totalInput += input;
154
+ stats.totalOutput += output;
155
+ stats.totalReasoning += reasoning;
156
+ stats.totalCacheRead += cacheRead;
157
+ stats.totalCacheWrite += cacheWrite;
158
+ stats.totalCost += cost;
159
+ stats.messageCount += 1;
160
+ // Log to file
161
+ logJson({
162
+ type: "tokens",
163
+ sessionId,
164
+ messageId,
165
+ role: info.role,
166
+ agent: info.agent,
167
+ model,
168
+ provider,
169
+ input,
170
+ output,
171
+ reasoning,
172
+ cacheRead,
173
+ cacheWrite,
174
+ cost,
175
+ });
176
+ // Show toast for this message
177
+ const totalTokens = input + output;
178
+ try {
179
+ await client.tui.showToast({
180
+ body: {
181
+ title: `${formatTokens(totalTokens)} tokens`,
182
+ message: `${formatCost(cost)} | Session: ${formatCost(stats.totalCost)}`,
183
+ variant: "info",
184
+ duration: 3000,
185
+ },
186
+ });
187
+ }
188
+ catch { }
189
+ }
190
+ // Handle session idle (show summary)
191
+ if (event.type === "session.idle") {
192
+ const props = event.properties;
193
+ const sessionId = props?.sessionID;
194
+ if (!sessionId)
195
+ return;
196
+ const stats = sessionStats.get(sessionId);
197
+ if (!stats || stats.messageCount === 0)
198
+ return;
199
+ const duration = Math.round((Date.now() - stats.startTime) / 1000 / 60);
200
+ const totalTokens = stats.totalInput + stats.totalOutput;
201
+ try {
202
+ await client.tui.showToast({
203
+ body: {
204
+ title: `Session: ${formatTokens(totalTokens)} tokens`,
205
+ message: `${formatCost(stats.totalCost)} | ${stats.messageCount} msgs | ${duration}min`,
206
+ variant: "info",
207
+ duration: 5000,
208
+ },
209
+ });
210
+ }
211
+ catch { }
212
+ }
213
+ }
214
+ catch { }
215
+ },
216
+ };
217
+ };
218
+ export default TokenTrackerPlugin;
package/package.json ADDED
@@ -0,0 +1,51 @@
1
+ {
2
+ "name": "opencode-token-tracker",
3
+ "version": "1.0.0",
4
+ "description": "Real-time token usage and cost tracking plugin for OpenCode with Toast notifications",
5
+ "main": "dist/index.js",
6
+ "types": "dist/index.d.ts",
7
+ "type": "module",
8
+ "scripts": {
9
+ "build": "tsc",
10
+ "prepublishOnly": "npm run build"
11
+ },
12
+ "keywords": [
13
+ "opencode",
14
+ "plugin",
15
+ "token",
16
+ "usage",
17
+ "cost",
18
+ "tracking",
19
+ "ai",
20
+ "llm",
21
+ "claude",
22
+ "gpt",
23
+ "anthropic",
24
+ "openai"
25
+ ],
26
+ "author": "tongsh6",
27
+ "license": "MIT",
28
+ "repository": {
29
+ "type": "git",
30
+ "url": "git+https://github.com/tongsh6/opencode-token-tracker.git"
31
+ },
32
+ "bugs": {
33
+ "url": "https://github.com/tongsh6/opencode-token-tracker/issues"
34
+ },
35
+ "homepage": "https://github.com/tongsh6/opencode-token-tracker#readme",
36
+ "dependencies": {
37
+ "@opencode-ai/plugin": "latest"
38
+ },
39
+ "devDependencies": {
40
+ "@types/node": "^20.0.0",
41
+ "typescript": "^5.0.0"
42
+ },
43
+ "files": [
44
+ "dist",
45
+ "README.md",
46
+ "LICENSE"
47
+ ],
48
+ "engines": {
49
+ "node": ">=18.0.0"
50
+ }
51
+ }