Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/riesbri/dshline/agents-mdgit clone --depth 1 https://github.com/riesbri/dshlineWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.05172 | $0.05172 |
| Opus 5 | $0.02586 | $0.02586 |
| Sonnet 5 | $0.01034 | $0.01034 |
| Haiku 4.5 | $0.00517 | $0.00517 |
Grade A, and why
dshline AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 243 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Instructions for an agent — or a person — working on this repository. If you want to use the interface, start at README.md. If you want to send a change, read CONTRIBUTING.md as well.
Read docs/design.md before changing anything about drawing, keyboard decoding, or text escaping. This file is the short version: the rules, the commands, and the mistakes that are easy to make.
What this project is
A terminal interface for DeepSeek Harness. It runs as a plugin inside the agent's own process, rather than as a client connecting over a network. There are two packages:
packages/renderer @dshline/renderer widths, keys, input line, boxes, screen — knows nothing about agents
packages/dshline @dshline/dshline the plugin: session loop, transcript, harness integration, view registry
packages/dshline/bin dshline a launcher wrapper, and deliberately nothing more
bin/dshline.mjs exists so that using this does not require remembering two things (dsh, --profile dshline). It finds the harness launcher, adds the profile and the working folder, and hands over the terminal with stdio: 'inherit'. It must never grow session logic: one implementation of the frontend is the point, and a wrapper that started doing its own work would be a second one.
That split is not just tidiness. The renderer must never import from the harness, and must never gain a dependency or a peer dependency. Having no dependencies is what lets this plugin add nothing to a user's setup, and it is why every rule below about widths, cutting, and escaping can be tested without a terminal and without a model.
Commands
pnpm install
pnpm build # tsc -b for both packages
pnpm test # the full suite, no terminal and no model required
pnpm typecheck # tsc -b, same project graph
pnpm security # the dependency and workflow checks CI runs
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 243 lines · 5,172 tokens per session scan A e5cb3daca1fb
dshline AGENTS.md is an instructions file published in the GitHub repository riesbri/dshline (11 stars, last pushed 4d ago), licensed MIT. It adds 5,172 tokens to every session, about $0.0259 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
dsh-web AGENTS.md
AGENTS.md instructions for zhu1090093659/dsh-web, covering dsh-web repository instructions, repository layout, common commands, repository rules and development workflow.
dsh-mcp-panel AGENTS.md
Instructions for PerryLink/dsh-mcp-panel, covering agents.md, layout, hard rules applied here, config and build.
dsh-codex-sync AGENTS.md
Instructions for Walvez/dsh-codex-sync, covering agents.md — dsh-codex-sync, 定位, 怎么跑, 技术栈 and 目录与约定.
dsh-industry-research AGENTS.md
Instructions for PerryLink/dsh-industry-research, covering agents.md, layout, hard rules applied here, checks and release.
dsh-mcp-manager AGENTS.md
Instructions for hyqhyq3/dsh-mcp-manager, covering agents.md, layout, architecture rules, commands and conventions.
omdp AGENTS.md
Instructions for XJungit/omdp, covering agents.md — omdp 仓库开发规范, 工作前提（先读再做）, 规范 1：文档随插件同步更新, 规范 2：教训经验主动沉淀为文档笔记 and 笔记索引.