dsh-token-usage
Run the following command in DeepSeek Harness:
dsh plugin install joyiok/dsh-token-usage
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install joyiok/dsh-token-usage in DeepSeek Harness to install this plugin; source code is available at https://github.com/joyiok/dsh-token-usage .
About this plugin
After a long day of DSH sessions, figuring out exactly how many tokens you have burned is surprisingly hard. dsh-token-usage puts the numbers right in front of you: a Tokens button appears in the bottom-left of the web sidebar, and one click reveals cumulative, today, and per-session input, output, and cache-read counts. No log files to scroll through.
It goes beyond a dashboard. The plugin registers a token_usage agent tool, so in headless or web sessions you can simply ask "how many tokens have I used in total" and get an authoritative answer. Scope switches between all, today, and current, and per-session detail is available on demand. Under the hood it reads session.jsonl (including .zstd) straight from $DSH_HOME, counts each turn-step exactly once, prefers assistant/message data.usage, and performs no character-level estimation. A built-in multi-frame zstd decoder means zero external binaries required.
If you run multi-turn DSH workloads, track API spend, or are simply curious about token output per session, this lightweight, non-intrusive, MIT-licensed plugin is a drop-in that pays off the moment you restart dsh web.
Use Cases
- One-click cumulative and daily token breakdown from the web sidebar
- Ask the agent directly in a headless session for an authoritative token total
- Inspect per-session token consumption across multi-turn DSH workloads
Best For
- Developers who need visibility into DSH session token spend
- Individuals or teams tracking API cost budgets
- Users running long multi-turn DSH workloads who need cost transparency
Related Plugins
Memory layer for coding agents that indexes local session history and auto-recalls relevant context before edits, commands, and failures, with no manual search needed.
Traceable, searchable cross-session memory for AI agents that turns conversation knowledge into a typed knowledge graph and recalls relevant subgraphs instead of replaying full history, natively integrated with DeepSeek Harness.
Gives DSH AI cross-session long-term memory, to-do and skill management, plus multi-session orchestration, external AI delegation, and an infinite canvas that grows with you.
dsh-mnemon is a three-tier, pluggable, Agent-driven memory system for DeepSeek Harness, combining Runtime memory, Project Documents, and replaceable Memory Spaces with nine long-term providers.