mirror of https://github.com/deepseek-ai/deepseek-harness.git synced 2026-08-29 04:26:38 +00:00

Files

T

Tianyi Cui e98c1c5d42 Add examples/coding-agent and the docs cookbook

The first real agent wiring: DeepSeek V4 + the bash tool suite + stdio
chat + JSONL persistence, runnable via yarn demo:coding (reads the
gitignored repo-root .env through process.loadEnvFile).

- examples/coding-agent: cordis.yml wiring both real plugin families
  (llm-deepseek with !!js env secrets; bash-local + tool-bash), a
  bash-only coding system prompt, a max-steps-guard plugin (bounds
  runaway turns via the agent/turn-continuation waterfall — abort()
  from step-end is a no-op by then), and a stdio UI with dimmed
  reasoning and exit-on-idle for piped stdin.
- e2e (yarn test:e2e, key-gated): full-loop.e2e.ts runs a real model
  against the real bash tool; coding-task.e2e.ts is the swebench-style
  smoke — the model fixes a buggy add.js in a temp dir and the test
  re-runs node add.test.js itself rather than trusting the agent.
- docs/cookbook: adding-a-package (the verified checklist),
  adding-a-tool (execute() contract, background pattern, seams),
  adding-an-llm-adapter (protocol obligations, mock-server testing,
  e2e policy). AGENTS.md layout/commands/secrets sections updated;
  architecture.md points at both examples and the cookbook.
- vitest.e2e.config.ts: serialize test files + retry twice — parallel
  e2e files trip the shared internal key's concurrency quota.
- fix: the !js YAML tag spelling in docs/JSDoc is actually !!js
  (js-yaml resolves custom tags under tag:yaml.org,2002:js).

2026-06-13 18:30:50 +08:00

3.2 KiB

Raw Blame History

Cookbook: adding a tool

How to give the model a new capability. Reference implementations: examples/echo-agent/src/echo-tool.ts (minimal) and packages/tool-bash (production-grade, three-package seam).

The minimal shape

import type { Context } from 'cordis'
import { defineTool } from '@deepseek-ai/dsh-tools'

export const name = 'my-tool'
export const inject = ['tools']

export function apply(ctx: Context) {
  ctx.tools.register(defineTool({
    name: 'read_file',
    description: 'Read a file from disk.',          // what the model sees
    parameters: {
      path: { type: 'string', required: true, description: 'Absolute path' },
      limit: { type: 'number' },                     // optional by default
    },
    async execute(args, exec) {
      // args is TYPED from the schema: { path: string; limit?: number }
      // exec carries { callId, name, arguments, agent?, signal? }
      return [{ type: 'text', text: await readFile(args.path, 'utf8') }]
    },
  }))
}

Registration is effect-based: disposing the plugin fiber unregisters the tool (write the HMR test). Schemas flow into the system-prompt assembly automatically.

Rules of the execute() contract

Validate args at runtime. defineTool's InferArgs typing is compile-time only; at runtime arguments is whatever JSON the model emitted. Check every field; throw a descriptive Error for bad input.
Throwing means isError. The registry catches anything execute() throws and returns {isError: true} to the model. Use that for infrastructure failures (bad input, spawn errors, aborts) — but REPORT domain failures in the result text instead (e.g. tool-bash returns [exit code: 9] with isError: false: the model decides what a failing command means).
Honor exec.signal. Cancel in-flight work when it fires.
Use exec.agent for async notifications. agent.inject(content, {source: {kind: 'plugin', plugin: '<name>'}}) appends durable context the NEXT model request sees — it is not a wake-up (an idle agent stays idle). Guard against disposed agents (try/catch).

Long-running work

Follow tool-bash's background pattern: a run_in_background flag returns a task id immediately; companion tools poll incrementally and kill; completion notices arrive via agent.inject(). Bound buffers and spill full output to disk so nothing is silently lost.

TODO: each tool reimplements this background pattern by hand today. At some point we need a generic long-running-tool layer that handles task ids, incremental polling, kill, and completion notices uniformly.

Permissions / sandboxing

Prefer not to build policy into the tool. The seam is the tools/execute waterfall (veto or wrap — see the permission-gate example in docs/architecture.md), or a sandboxing implementation behind the tool's executor seam.

Tests every tool needs

Arg-validation rejections, result shaping for every outcome, the HMR disposal test, and — for tools with side effects — an integration spec that drives the tool through the agent loop with a scripted MockAdapter (packages/agent-loop/tests/mock-adapter.ts), asserting the tool/call / tool/result session events.

3.2 KiB Raw Blame History