Pine/deepseek-harness
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases 1 Wiki Activity

feat(goal): add human goal command

Browse Source
This commit is contained in:
Tianyi Cui
2026-07-19 23:55:33 +08:00
parent 400b4658ef
commit 207692bc16
114 changed files with 1831 additions and 88 deletions
Download Patch File Download Diff File Expand all files Collapse all files
packages/examples/agent-spine-demo/README.md
+7 -4
View File
@@ -17,6 +17,9 @@ Read this package for the whole plugin tree and its composition order.
@deepseek-ai/dsh-skill skill provider registry
@deepseek-ai/dsh-skill-local local filesystem skill provider
@deepseek-ai/dsh-agent agent registry + initiator scope + agent/* events
@deepseek-ai/dsh-goal optional persisted same-session goal domain
@deepseek-ai/dsh-tool-goal optional model-facing goal controls
@deepseek-ai/dsh-goal-session optional same-session goal-round driver
@deepseek-ai/dsh-tasks generic background-task registry
@deepseek-ai/dsh-invariants dev-mode event-contract assertions
@deepseek-ai/dsh-tool-bash the model-facing bash schema
@@ -42,11 +45,11 @@ This is the [interface/implementation/consumer seam](../../../docs/rfc/implement
```ts
import type { Config } from '@deepseek-ai/dsh-agent-spine-demo'
// { agents?, maxParallelToolCalls?, persona?, toolOrder?, tools?, dshHome?, skills?, workspaceContext, toolBash?, toolTasks? }
// { agents?, maxParallelToolCalls?, persona?, toolOrder?, tools?, dshHome?, skills?, workspaceContext, toolBash?, toolTasks?, goals? }
// workspaceContext requires { maxBytes } or false; the other owner schemas supply defaults.
```
The bundle FORWARDS each field to the child that owns it: `agents` and `maxParallelToolCalls` to `agent-loop` (`agents` defaults to `[]`; the cap defaults there), so each app supplies its own pre-created agents — a stdio app pre-creates `main`, while the ACP app creates agents on demand at `session/new`; `persona` and `toolOrder` to `dsh-system-prompt`; `tools` to the tool registry for its presentation mode; `skills.registry`, `skills.local`, and `skills.tool` to the skill registry, local provider, and model-facing consumer; the required `workspaceContext` choice to `dsh-workspace-context` (`{ maxBytes }` enables loading and `false` disables it); and `toolBash`/`toolTasks` to the two model-facing tool plugins the bundle owns. Set `skills.enabled: false` to omit both the local provider and model-facing skill tool, and set `toolTasks: false` to retain the task service for foreground producers without exposing `task_output` / `task_list` / `task_kill`. It resolves `dshHome` once through [`@deepseek-ai/dsh-home`](../../util/home/README.md) and forwards that absolute value to tool-bash's managed environment and enabled local skill discovery. An absent top-level `dshHome` adopts `skills.local.dshHome`; supplying both with different resolved paths fails loudly. `toolBash.enableRunInBackground` controls only the bash producer; independently loaded producers keep their own config. Workspace instructions register before the skill catalog so their session-prefix message renders first. App packages use `pickSpineConfig()` to copy only these bundle-owned fields.
The bundle FORWARDS each field to the child that owns it: `agents` and `maxParallelToolCalls` to `agent-loop` (`agents` defaults to `[]`; the cap defaults there), so each app supplies its own pre-created agents — a stdio app pre-creates `main`, while the ACP app creates agents on demand at `session/new`; `persona` and `toolOrder` to `dsh-system-prompt`; `tools` to the tool registry for its presentation mode; `skills.registry`, `skills.local`, and `skills.tool` to the skill registry, local provider, and model-facing consumer; the required `workspaceContext` choice to `dsh-workspace-context` (`{ maxBytes }` enables loading and `false` disables it); and `toolBash`/`toolTasks` to the two model-facing tool plugins the bundle owns. A `goals` object opts into the persisted domain, model tools, and same-session driver while forwarding `goals.domain` and `goals.tool` to their owners; omission or `false` leaves the stack absent so headless callers retain one-turn settlement. Set `skills.enabled: false` to omit both the local provider and model-facing skill tool, and set `toolTasks: false` to retain the task service for foreground producers without exposing `task_output` / `task_list` / `task_kill`. It resolves `dshHome` once through [`@deepseek-ai/dsh-home`](../../util/home/README.md) and forwards that absolute value to tool-bash's managed environment and enabled local skill discovery. An absent top-level `dshHome` adopts `skills.local.dshHome`; supplying both with different resolved paths fails loudly. `toolBash.enableRunInBackground` controls only the bash producer; independently loaded producers keep their own config. Workspace instructions register before the skill catalog so their session-prefix message renders first. App packages use `pickSpineConfig()` to copy only these bundle-owned fields.
## Why a code bundle, not a shared YAML include
@@ -54,7 +57,7 @@ A YAML include can deduplicate config but cannot own a bin or provide front-door
## Model Experience
Indirectly, through `dsh-system-prompt`, `dsh-tool-skill`, `dsh-tool-bash`, and `dsh-tools`, which this bundle mounts without adding model-bound wrapper content.
Indirectly, through `dsh-system-prompt`, `dsh-tool-skill`, `dsh-tool-bash`, and `dsh-tools`, plus `dsh-tool-goal` and goal-round prompts when `goals` is enabled. The bundle adds no model-bound wrapper content of its own.
#### KV Cache effect
@@ -62,5 +65,5 @@ No direct invalidation; the named consumer owns any request-prefix changes.
## Known Limitations and Deferred Work
- **Most of the spine set is fixed in code** — `apply()` always mounts the core services and `tool-bash`; config can omit the bundled skills and task-control tools, but swapping the loop or dropping another spine member means composing a different bundle.
- **Most of the spine set is fixed in code** — `apply()` always mounts the core services and `tool-bash`; config can omit bundled goals, skills, and task-control tools, but swapping the loop or dropping another spine member means composing a different bundle.
- **`dsh-invariants` mounts unconditionally** — this bundle has no toggle, so every composition using it pays the dev-mode relational assertions; Session's always-on validation and freezing are separate.
packages/examples/agent-spine-demo/package.json
+7 -1
View File
@@ -1,6 +1,6 @@
{
"name": "@deepseek-ai/dsh-agent-spine-demo",
"description": "The default executor-less/UI-less agent spine as one Cordis bundle plugin (timer + llm + sessions + system-prompt + tools + skills + agents + tasks + invariants + tool-bash + workspace-context + tool-skill + tool-tasks + agent-loop)",
"description": "The default executor-less/UI-less agent spine as one Cordis bundle plugin with optional persisted goals",
"version": "0.0.1",
"private": true,
"type": "module",
@@ -25,6 +25,8 @@
"@cordisjs/plugin-timer": "^1.1.2",
"@deepseek-ai/dsh-agent": "^0.0.1",
"@deepseek-ai/dsh-agent-loop": "^0.0.1",
"@deepseek-ai/dsh-goal": "^0.0.1",
"@deepseek-ai/dsh-goal-session": "^0.0.1",
"@deepseek-ai/dsh-invariants": "^0.0.1",
"@deepseek-ai/dsh-home": "^0.0.1",
"@deepseek-ai/dsh-llm": "^0.0.1",
@@ -35,6 +37,7 @@
"@deepseek-ai/dsh-system-prompt": "^0.0.1",
"@deepseek-ai/dsh-tasks": "^0.0.1",
"@deepseek-ai/dsh-tool-bash": "^0.0.1",
"@deepseek-ai/dsh-tool-goal": "^0.0.1",
"@deepseek-ai/dsh-tool-skill": "^0.0.1",
"@deepseek-ai/dsh-tool-tasks": "^0.0.1",
"@deepseek-ai/dsh-tools": "^0.0.1",
@@ -44,6 +47,8 @@
"@cordisjs/plugin-timer": "workspace:^",
"@deepseek-ai/dsh-agent": "workspace:^",
"@deepseek-ai/dsh-agent-loop": "workspace:^",
"@deepseek-ai/dsh-goal": "workspace:^",
"@deepseek-ai/dsh-goal-session": "workspace:^",
"@deepseek-ai/dsh-fs-local": "workspace:^",
"@deepseek-ai/dsh-invariants": "workspace:^",
"@deepseek-ai/dsh-home": "workspace:^",
@@ -55,6 +60,7 @@
"@deepseek-ai/dsh-system-prompt": "workspace:^",
"@deepseek-ai/dsh-tasks": "workspace:^",
"@deepseek-ai/dsh-tool-bash": "workspace:^",
"@deepseek-ai/dsh-tool-goal": "workspace:^",
"@deepseek-ai/dsh-tool-skill": "workspace:^",
"@deepseek-ai/dsh-tool-tasks": "workspace:^",
"@deepseek-ai/dsh-tools": "workspace:^",
packages/examples/agent-spine-demo/src/index.ts
+30 -3
View File
@@ -1,6 +1,6 @@
/**
* Default executor-less, UI-less agent spine. It bundles the common services,
* background-task registry and controls, concrete loop, local skill and
* background-task registry and controls, optional persisted goals, concrete loop, local skill and
* workspace-context providers, and model-facing bash/skill consumers;
* deployments still choose the LLM adapter, bash executor, and presentation.
* The plugin intentionally exposes named exports only because Loader default
@@ -18,6 +18,9 @@ import ToolRegistry, { type Config as ToolsConfig } from '@deepseek-ai/dsh-tools
import SkillService, { type Config as SkillRegistryConfig } from '@deepseek-ai/dsh-skill'
import * as SkillLocal from '@deepseek-ai/dsh-skill-local'
import AgentRegistry from '@deepseek-ai/dsh-agent'
import GoalService, { type Config as GoalDomainConfig } from '@deepseek-ai/dsh-goal'
import * as goalSession from '@deepseek-ai/dsh-goal-session'
import * as toolGoal from '@deepseek-ai/dsh-tool-goal'
import TaskService from '@deepseek-ai/dsh-tasks'
import * as invariants from '@deepseek-ai/dsh-invariants'
import * as toolBash from '@deepseek-ai/dsh-tool-bash'
@@ -41,6 +44,14 @@ export interface SkillConfig {
tool?: toolSkill.Config
}
/** Persisted goal domain, model-tool policy, and same-session driver config. */
export interface GoalConfig {
/** Goal-domain creation defaults. */
domain?: GoalDomainConfig
/** Model-facing goal-tool authority policy. */
tool?: toolGoal.Config
}
/**
* Bundle config: each field forwarded verbatim to the child that owns it —
* `agents` to the agent loop (an app that pre-creates no agents, like the ACP
@@ -50,7 +61,8 @@ export interface SkillConfig {
* `dshHome` to bash environment and local skill discovery, `skills` to the
* skill registry/local provider/tool consumer, `workspaceContext` to the
* workspace-context loader, and `toolBash`/`toolTasks` to the model-facing tool
* plugins this bundle owns. Owner schemas supply defaults for optional input;
* plugins this bundle owns. `goals` opts into and configures the persisted goal
* domain plus its model tool and same-session driver. Owner schemas supply defaults for optional input;
* workspace context instead requires an explicit byte budget or `false` because
* it changes model-visible input. Producer opt-in stays producer-local:
* `toolBash` configures bash only; independently composed producers keep their
@@ -77,6 +89,8 @@ export interface Config {
toolBash?: toolBash.Config
/** Generic background-task controls; set false to keep the task service without model-facing task tools. */
toolTasks?: toolTasks.Config | false
/** Opt-in persisted same-session goal stack; set false or omit to leave it unmounted. */
goals?: GoalConfig | false
}
/** The skill config schema exported for app packages that forward `skills`. */
@@ -93,6 +107,12 @@ export const ToolBashConfigSchema: z<toolBash.Config> = toolBash.Config
/** The task-control-tool config schema exported for app packages that forward `toolTasks`. */
export const ToolTasksConfigSchema: z<toolTasks.Config> = toolTasks.Config
/** The persisted-goal config schema exported for app packages that opt in. */
export const GoalConfigSchema: z<GoalConfig> = z.object({
domain: GoalService.Config,
tool: toolGoal.Config,
})
/** Intersect the owners' schemas so validation + defaulting stay identical. */
export const Config = z.intersect([
AgentLoop.Config,
@@ -104,7 +124,8 @@ export const Config = z.intersect([
workspaceContext: z.union([z.const(false), workspaceContext.Config]).required(),
toolBash: ToolBashConfigSchema,
toolTasks: z.union([z.const(false), ToolTasksConfigSchema]),
}) as unknown as z<Pick<Config, 'tools' | 'dshHome' | 'skills' | 'workspaceContext' | 'toolBash' | 'toolTasks'>>,
goals: z.union([z.const(false), GoalConfigSchema]),
}) as unknown as z<Pick<Config, 'tools' | 'dshHome' | 'skills' | 'workspaceContext' | 'toolBash' | 'toolTasks' | 'goals'>>,
]) as unknown as z<Config>
/**
@@ -123,6 +144,7 @@ export function pickSpineConfig(config: Omit<Config, 'agents'>): Omit<Config, 'a
...config.skills !== undefined ? { skills: config.skills } : {},
...config.toolBash !== undefined ? { toolBash: config.toolBash } : {},
...config.toolTasks !== undefined ? { toolTasks: config.toolTasks } : {},
...config.goals !== undefined ? { goals: config.goals } : {},
}
}
@@ -159,6 +181,11 @@ export function apply(ctx: Context, config: Config): void {
ctx.plugin(SkillLocal, Object.assign({}, config.skills?.local, { dshHome }))
}
ctx.plugin(AgentRegistry)
if (config.goals !== undefined && config.goals !== false) {
ctx.plugin(GoalService, config.goals.domain ?? {})
ctx.plugin(toolGoal, config.goals.tool ?? {})
ctx.plugin(goalSession)
}
ctx.plugin(TaskService)
ctx.plugin(invariants)
ctx.plugin(toolBash, Object.assign({}, config.toolBash, { dshHome }))
packages/examples/agent-spine-demo/tests/agent-core.spec.ts
+39
View File
@@ -114,6 +114,33 @@ describe('dsh-agent-spine-demo bundle', () => {
expect(ctx.get('agents')).toBeDefined()
expect(ctx.get('tasks')).toBeDefined()
expect(ctx.get('agentLoop')).toBeDefined()
expect(ctx.get('goals')).toBeUndefined()
await ctx.fiber.dispose()
})
it('opts into the configured persisted-goal domain, tools, and same-session driver', async () => {
const ctx = await mount({
workspaceContext: false,
goals: {
domain: { defaultMaxGoalRounds: 17 },
tool: { blockedAfterConsecutiveRounds: 5 },
},
})
expect(ctx.goals.resolveCreate({ objective: 'configured' })).toEqual({
objective: 'configured',
maxGoalRounds: 17,
})
expect(['create_goal', 'get_goal', 'update_goal'].map(name => ctx.tools.get(name)?.name))
.toEqual(['create_goal', 'get_goal', 'update_goal'])
expect((await ctx.systemPrompt.assemble()).sections.find(section => section.name === 'tool:goal')?.text)
.toContain('at least 5 consecutive goal rounds')
await ctx.fiber.dispose()
})
it('accepts an explicit false goal composition without mounting it', async () => {
const ctx = await mount({ workspaceContext: false, goals: false })
expect(ctx.get('goals')).toBeUndefined()
expect(ctx.tools.get('get_goal')).toBeUndefined()
await ctx.fiber.dispose()
})
@@ -170,6 +197,18 @@ describe('dsh-agent-spine-demo bundle', () => {
await ctx.fiber.dispose()
})
it('uses owner defaults for a schema-bypassing empty goal opt-in', async () => {
const ctx = new Context()
agentCore.apply(ctx, { workspaceContext: false, goals: {} })
await new Promise(resolve => setTimeout(resolve, 50))
expect(ctx.goals.resolveCreate({ objective: 'defaulted' })).toEqual({
objective: 'defaulted',
maxGoalRounds: 256,
})
expect(ctx.tools.get('get_goal')).toBeDefined()
await ctx.fiber.dispose()
})
it('loads workspace instructions into requests through the bundled spine', async () => {
const root = await mkdtemp(join(tmpdir(), 'dsh-agent-spine-demo-workspace-context-'))
try {
packages/examples/agent-spine-demo/tsconfig.json
+9
View File
@@ -41,6 +41,15 @@
{
"path": "../../core/agent"
},
{
"path": "../../goal/goal"
},
{
"path": "../../goal/tool-goal"
},
{
"path": "../../goal/goal-session"
},
{
"path": "../../context/workspace-context"
},