@qiaoy01/mission
DeepSeek Harness 的持久化长时域自主任务运行时：依赖 DAG、独立验证、失败时重新规划、租约所有权。
插件会安装到这里；不确定时保持 web。
npx -y @deepseek-ai/dsh plugin --profile web add @qiaoy01/mission@0.2.0Mission 以 @qiaoy01/mission 发布，当前版本为 0.2.0。Plugin Hub 会校验它的 manifest，并保存精确安装来源，便于复现安装结果。
A long-horizon autonomous mission runtime for DeepSeek Harness — lets an agent drive a long-term objective over days: a cross-session, independently verifiable, replan-on-failure, crash-recoverable autonomous control loop. The dependency DAG is one of its driving mechanisms, not the runtime itself.
English | 中文
dsh-mission is a pure-TypeScript cordis plugin for DeepSeek Harness that
provides the long-horizon autonomous mission runtime dsh lacks: a durable,
verifiable, crash-recoverable control loop that breaks a long-term objective
into a dependency DAG of tasks and advances it autonomously — instead of
finishing everything in one conversation.
Core principle: agents propose, the environment adjudicates, the runtime commits.
task/report is a declaration about the environment
(not authoritative); task/verify independently aligns the declaration
with environment feedback (authoritative);| What dsh already has | What dsh-mission adds |
|---|---|
| goal (single-objective ledger, caller self-certifies) | Dependency DAG + derived readiness/blocked, failed branches replan while DONE work is preserved |
| workflow (one-shot fan-out scripts) | Durable, resumable, cross-session task state machine |
| schedule (pure time triggers) | Dependency triggers (upstream DONE → downstream READY) |
| subagent (execution unit) | Independent verification — workers cannot self-certify, report ≠ verify |
| — | Lease ownership + lazy reclamation + crash recovery |
mission/* session events; one event = one
atomic commit; no separate database;revision CAS (mirrors dsh-goal's GoalRef pattern);mission-driver): listens to mission/changed
and runs claim → subagent execute → report → verify → commit automatically;llmDecider (pick the next task) /
llmReplanner (replan) / userApprovalGate (approval gate);mission session-projection unit;mission-trigger): a soft-guidance prompt + /mission
slash command that routes trigger phrases into mission_create/mission_plan;mission-invariant registers its audit whenever the
invariants service exists, and is a no-op otherwise (production profiles
load without extra configuration).mission/*)| Event | Meaning |
|---|---|
mission/created | Create a mission |
mission/plan-committed | Commit a full plan snapshot (revision +1; DONE tasks with unchanged specs carry over) |
mission/task-claimed | Claim a task with a lease |
mission/task-reported | The agent's declaration about the environment (never DONE by itself) |
mission/task-verified | Independently align the declaration with environment feedback → DONE/FAILED |
mission/cancelled | Explicit cancellation |
States: mission CREATED / RUNNING / COMPLETED / FAILED / CANCELLED;
tasks PENDING / CLAIMED / DONE / FAILED + derived READY / BLOCKED.
# Install from npm (published; inserts a 6-row bundle patch into the composition)
dsh plugin --profile <name> add @qiaoy01/mission
The published npm package ships prebuilt (lib/), so consumers only need the
one command above. Build from source when you want to modify or review the code:
git clone https://github.com/qiaoy01/dsh-mission.git
cd dsh-mission # the bundle lives at this repo root
npm install # build deps, pinned to the running-harness peer versions
npm run build # typecheck + emit lib/
dsh plugin --profile <name> add file:.
file:.points at the repo root you justcd'd into — whosepackage.jsondeclaresdsh.bundle.patch→cordis.patch.yml. After changing source, re-runnpm run buildand re-add the package.
In a session, the agent drives the mission through four model tools:
mission_create (create a mission), mission_plan (commit a task DAG; revision
must equal current + 1), mission_status (read the current state), and
mission_cancel (cancel the mission).
After the model commits a plan, the runtime takes over: mission-driver
claims tasks, dispatches real subagents, and verifies independently. The model
should not execute tasks itself.
# 1. install the plugin into a dsh profile
dsh plugin --profile web add @qiaoy01/mission
# 2. in a session, ask the agent:
# "mission: build a small web game. Use mission_create, then mission_plan,
# then stop — the runtime will execute the tasks."
#
# The model creates and plans; mission-driver claims tasks, dispatches real
# subagents, and independently verifies each one until the mission is COMPLETED.
Defaults are conservative (deterministic task selection, no automatic replan — zero model cost). Enable model strategies via the profile row config:
- id: mission-driver
config:
decider: llm # the model picks the next task
replan: true # the model replans when the mission is stuck
npm run build # auto-discovers a local tsc; typecheck + emit to lib/
# Six test suites, all run in isolation:
node --experimental-strip-types tests/domain.test.mjs # pure domain model
node tests/service.test.mjs # runtime (real cordis/session/agent)
node tests/driver.test.mjs # driver loop (mock executor)
node tests/driver-host.test.mjs # driver host (stubbed subagents)
node tests/trigger.test.mjs # trigger companion (soft guidance + /mission)
node --experimental-strip-types tests/projection.test.mjs # session-projection unit
MIT — see LICENSE.