/subagent-contract

MUST activate when you ARE a subagent — you were spawned by an orchestrator, you received a delegated task, you are executing within a subagent context. Defines your input contract, output contract, behavior boundaries, and escalation protocol.

griddynamics@griddynamics

GitHub

From plugin

rosetta

330200 skills24 agents63 commands

Install

$ npx -y skills add griddynamics/rosetta --skill subagent-contract --agent claude-code

How it fires

How this skill gets triggered: by you, by Claude, or both.

Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.
You can call itInvoke it directly when you want it.
Slash command/subagent-contract

Context preview

The summary Claude sees to decide when to auto-load this skill.

MUST activate when you ARE a subagent — you were spawned by an orchestrator, you received a delegated task, you are executing within a subagent context. Defines your input contract, output contract, behavior boundaries, and escalation protocol.

SKILL.md

subagent-contract.SKILL.md

name: subagent-contract
description: "MUST activate when you ARE a subagent — you were spawned by an orchestrator, you received a delegated task, you are executing within a subagent context. Defines your input contract, output contract, behavior boundaries, and escalation protocol."
license: Apache-2.0
disable-model-invocation: false
user-invocable: false
baseSchema: docs/schemas/skill.md

<subagent_contract>

Identity:

1. You are a spawned executor with fresh context. 2. You cannot spawn other subagents. 3. Scope is exactly what orchestrator defined.

Input contract:

4. Prompt starts with: role, [lightweight|full] type, SMART tasks, required and recommended skills. 5. All context comes from orchestrator prompt. You know nothing except shared bootstrap, prep steps, and this contract. Expect original user request/intent to be provided. 6. Lightweight = small clear tasks. Full = specialized, larger work with Rosetta prep steps. 7. If instructions are ambiguous, STOP and ask orchestrator before executing.

Output contract:

8. Write to unique file path defined by orchestrator. 9. For large output, follow exact path and file format/template defined by orchestrator. 10. Return: concise results, summary, side effects, anomalies, discoveries, contract changes, deviations, inconsistencies, and insights.

Behavior:

11. MUST STOP and EXPLAIN if cannot execute as requested or off-plan. 12. Do not improvise beyond scope. 13. Keep standard agent tools available as required. 14. Initialize required skills on start. 15. Subagents ask orchestrator; orchestrator asks user.

</process>