dsh-plugin-subagent-roster

dsh-plugin-subagent-roster

A constrained subagent roster for DeepSeek Harness. Every role has a pinned system prompt, a pinned capability range, and a pinned model tier — and bulk work and team work are reached through tools of different shapes.

The problem

DSH's built-in subagent tool gives a child the parent's identity and the parent's whole toolset. Real delegation therefore fails in three predictable ways:

1. Identity leaks — the child carries the parent's persona into a narrow job, and its output shape is whatever the model felt like.
2. Capability leaks — you wanted a file read; the child has a shell. You asked for a review; it edited the code.
3. Model does not match the job — a 200-item extraction and a load-bearing design decision run on the same (usually most expensive) model.

This plugin turns all three from prompt requests into structural constraints:

Constraint	Mechanism	Effect
System prompt	each delegation puts the role persona on the request; it becomes the child's persona-prefix section	identity is decided by the role, never by the parent
Capability	the request's toolFilter.allow	a reviewer has no write tool at all — it cannot edit even if asked
Model	the request's agentOptions	bulk runs cheap, review runs adequate, team runs strongest
Depth	maxDepth: 1 and a filter that never grants subagent*	a role child is a leaf; no recursive fan-out

Two lines

• Bulk line — one item per call, launched in parallel, children have no tools at all. Field extraction over 20 records, per-item validation, per-item rewriting.
• Team line — planning, building, reviewing, retrieving, assembling. Each role has a different capability range, and no role can stand in for another.

All 20 roles: role reference (中文).

Install

# from GitHub (source is the artifact; no build authorization needed)
dsh plugin --profile  add github:/dsh-plugin-subagent-roster

# or from a local checkout
dsh plugin --profile  add /path/to/dsh-plugin-subagent-roster

The bundle applies one plugin row. The parent agent gains one delegate tool whose role enum is the whole roster, so it pays for one schema instead of twenty. dsh --profile --dump-config should show the subagent-roster row.

Operations page (define your own roles)

While a profile runs, the plugin serves a loopback page:

http://127.0.0.1:4317/

Add or delete roles, change a role's position, temperament, tier or capability range, and edit the persona — the child's system prompt — directly. Saving applies immediately, no restart: the page replaces the delegation tool registration and the routing section, and rolls back to the previous roster if the new one fails to mount.

Roles live in $DSH_HOME/subagent-roster/roles.json, seeded from the shipped twenty on first start. The file is a normal document: editing it by hand is supported.

• Port is the plugin's pagePort setting; 0 disables the page.
• No authentication, bound to 127.0.0.1: it grants the same file authority a local shell already has. Do not expose it on another interface.
• Every save is validated as a whole: kebab-case ids, known tier/capability/line values, and role invariants (a reviewer that can write is rejected). A rejected document never overwrites the file.

Deterministic entry point (no "should I delegate?" decision)

Measured: letting the parent model decide whether to delegate is unreliable on free models (see test report §3). So the decision moves to the entry point — drive it by item count:

node bin/roster-fanout.mjs \
  --profile  \
  --role batch-exact \
  --items-file items.json \
  --contract 'one JSON object {"text": }' \
  --expect-json \
  --concurrency 3 \
  --out results.jsonl

• N items in, N role-children out — the shape is a property of the invocation, not the model
• Each answer is read from the child's tool_result event, never from the parent's paraphrase
• --expect-json validates the contract; a child that errors, returns nothing, or returns invalid JSON fails the item
• Non-zero exit when any item failed, so a script or CI job can gate on it

The contract is enforced by --expect-json, and the contract text is delivered inside the child's task (outside it, the child never sees it). Measured 8/8 compliant on the cheapest bulk-tier model. A failed check is a failed item, never prose passed off as success.

Plugin market (official index format)

The repository ships a format: 1 index for the DSH plugin market, validated with the official @deepseek-ai/dsh-plugin-market-index validator. Subscribe to it directly:

https://121212165.github.io/dsh-plugin-subagent-roster/market/index.json

Plugins → Discover → Add index source, paste the URL (turn on “Custom plugin market” under “Experimental plugins” first).

The index lists one version from a commit-pinned git source (24f55f2), so the source shown in the confirmation dialog is exactly what installs, and it cannot drift with a branch.

The repo also carries the dsh-plugin topic, so topic-scraping aggregators (awesome-dsh-plugin and friends) pick it up on their next crawl.

Configure model tiers

Every model id in the composition resolves from the environment, so one package serves several deployments:

Variable	Purpose
ROSTER_PROVIDER	provider route shared by all four tiers
ROSTER_BATCH_MODEL / ROSTER_BUILD_MODEL / ROSTER_REVIEW_MODEL / ROSTER_TEAM_MODEL	model id per tier
ROSTER_BATCH_MAX_TOKENS and friends	output ceiling per tier

Any single row can also be overridden by id in a profile's cordis.patch.yml.

Plugin settings

Field	Default	Meaning
mode	advisory	advisory warns; strict refuses to load a self-contradicting roster
provider	deepseek-official	provider the tiers resolve against
allowedModels	[]	when non-empty, any tier resolving outside this list is a violation
routingGuidance	true	register the bulk-vs-team routing section into the parent prompt

Verify

npm test                          # 18 unit tests, zero model calls
node bench/run.mjs --repeat 1     # real runs against a local OpenAI-compatible gateway

Bench verdicts come from the gateway's request log, not from answer text: which persona the child actually ran with, whether its model equals the tier model, whether its tool count equals the role's grant, and whether negative cases really left the workspace untouched.

• Design notes: docs/design.md
• Test report: docs/test-report.md
• Review: docs/review.md

Compatibility

• DSH >=0.2.0-rc.2 =22.18
• Plain ESM JavaScript artifact, no build step

License

MIT