Delegation Core Framework

Overview

A method for deciding when and how to delegate tasks to external LLM services. Core principle: delegate execution, retain high-level reasoning.

Default Posture

Delegation is on. Declining it is the step that takes a decision.

Any conjure operation, mission phase, or workflow that reaches eligible work delegates it without being asked. Eligible means the work is execution rather than reasoning, which the philosophy below defines and modules/task-assessment.md classifies. An operator who wants delegation off says so once, in one of the two places under Declining Delegation.

This inverts the earlier framing, which listed the occasions on which to delegate and so made delegation something a caller had to remember. The occasions were correct and are now the exception list instead.

Keep Local

Delegation stops at the boundary the philosophy draws. Keep the work local when any of these hold:

• The task is reasoning: architecture, design, trade-offs, judgment.
• The prompt or context carries a secret, a credential, or data that should not leave the machine.
• The task needs iteration with the operator turn by turn.
• Correctness cannot be validated after the fact.

The first is the standing exception and covers most of what Claude does. The rest are the red flags modules/task-assessment.md already lists.

Declining Delegation

Two switches, environment over file.

Scope	How	Effect
One run	CONJURE_DELEGATION=off	No provider is probed or spawned
One machine	"enabled": false in ~/.claude/hooks/delegation/config.json	Same, until the key changes

CONJURE_DELEGATION accepts off, 0, false, no and their opposites, in any case. Setting it to an on spelling re-enables delegation that the config file turned off, so the narrower scope wins. A disabled delegator returns a result whose fallback_reason is delegation_disabled and spawns nothing, so opting out costs nothing.

When No Provider Answers

smart_delegate works down the registry order and returns the first real answer. A provider that is missing, unauthenticated, failing, or answering with an empty stdout is recorded and the chain moves on.

When it reaches the end, it returns a result with fallback_reason set to providers_exhausted and an attempts trail naming what each provider did. That result is the instruction to do the work locally. Read it, say which providers were tried and why none answered, and then do the task yourself. Do not treat it as an error to report and stop on: a machine with no CLI installed is the ordinary case for this default, not a fault.

When the operator wants providers answering rather than falling back, modules/provider-onboarding.md carries the per-provider steps and the failure shapes that do not look like authentication problems. Start with make -C plugins/conjure delegate-doctor.

Philosophy

Delegate execution, retain reasoning. Claude handles architecture, strategy, design, and review. External LLMs perform data processing, pattern extraction, bulk operations, and summarization.

Delegation Flow

1. Task Assessment: Classify task by complexity and context size.
2. Suitability Evaluation: Check prerequisites and service fit.
3. Handoff Planning: Formulate request and document plan.
4. Execution & Integration: Run delegation, validate, and integrate results.

Quick Decision Matrix

Complexity	Context	Recommendation
High	Any	Keep local
Low	Large	Delegate
Low	Small	Either

High Complexity: Architecture, design decisions, trade-offs, creative problem solving.

Low Complexity: Pattern counting, bulk extraction, boilerplate generation, summarization.

Detailed Workflow Steps

1. Task Assessment (delegation-core:task-assessed)

Classify the task:

• See modules/task-assessment.md for classification criteria.
• Use token estimates to determine thresholds.
• Apply the decision matrix.

Exit Criteria: Task classified with complexity level, context size, and delegation recommendation.

2. Suitability Evaluation (delegation-core:delegation-suitability)

Verify prerequisites:

• See modules/handoff-patterns.md for checklist.
• Evaluate cost-benefit ratio using modules/cost-estimation.md.
• Check for red flags (security, real-time iteration).

Exit Criteria: Service authenticated, quotas verified, cost justified.

3. Handoff Planning (delegation-core:handoff-planned)

Create a delegation plan:

• See modules/handoff-patterns.md for request template.
• Document service, command, input context, expected output.
• Define validation method.

Exit Criteria: Delegation plan documented.

4. Execution & Integration (delegation-core:results-integrated)

Execute and validate results:

• Run delegation and capture output.
• Validate format and correctness.
• Integrate only after validation passes.
• Log usage.

Exit Criteria: Results validated and integrated, usage logged.

MCP Authentication

OAuth Client Credentials (Claude Code 2.1.30+)

For MCP servers that don't support Dynamic Client Registration (e.g., Slack), pre-configured OAuth client credentials can be provided:

claude mcp add <server-name> --client-id <id> --client-secret <secret>

This enables delegation workflows through MCP servers that require pre-configured OAuth, expanding the range of external services available for task delegation.

Claude.ai MCP Connectors (Claude Code 2.1.46+)

As an alternative to manual OAuth setup, users can configure MCP servers directly in claude.ai at claude.ai/settings/connectors. These connectors are automatically available in Claude Code when logged in with a claude.ai account: no claude mcp add or credential management required. This provides a browser-based auth flow that may be simpler for services with complex OAuth requirements.

Worktree Isolation for File-Modifying Delegations (Claude Code 2.1.49+)

When delegating tasks that modify files to subagents, use isolation: worktree in the agent frontmatter to run each agent in a temporary git worktree. This prevents file conflicts when multiple delegated agents operate in parallel on overlapping paths. The worktree is auto-cleaned if no changes are made; preserved with commits if the agent produces changes.

# Agent frontmatter for isolated delegation
isolation: worktree

Leyline Infrastructure

Conjure uses leyline infrastructure:

Leyline Skill	Used For
quota-management	Track service quotas and thresholds.
usage-logging	Session-aware audit trails.
service-registry	Unified service configuration.
error-patterns	Consistent error handling.
authentication-patterns	Auth verification.

See modules/cost-estimation.md for leyline integration examples.

Service-Specific Skills

For detailed service workflows:

• Skill(conjure:gemini-delegation): Gemini CLI specifics.
• Skill(conjure:qwen-delegation): Qwen MCP specifics.
• Skill(conjure:minimax-delegation): MiniMax CLI specifics.
• Skill(conjure:glm-delegation): GLM via the Z.ai endpoint swap.
• Skill(conjure:muse-delegation): Meta Muse Code CLI specifics.
• Skill(conjure:codex-delegation): OpenAI Codex CLI specifics.
• Skill(conjure:opencode-delegation): OpenCode CLI specifics.

Execution Modes

When delegating to multiple agents, choose the appropriate execution mode:

Mode	When to Use	How It Works
single-session	Sequential tasks, same-file edits	Claude works through tasks in order
subagents	Parallel independent tasks	Agents work independently, report back
agent-team	Parallel coordinated tasks	Agents can communicate with each other

See references/execution-modes.md for the selection decision matrix, mode compatibility notes, and anti-patterns to avoid.

Module Reference

• provider-onboarding.md: Install and auth per provider, and how to read a half-configured one.
• task-assessment.md: Complexity classification, decision matrix.
• cost-estimation.md: Pricing, budgets, cost tracking.
• handoff-patterns.md: Request templates, workflows.
• troubleshooting.md: Common problems, service failures.

Exit Criteria

• Task assessed and classified.
• Work that stayed local names which Keep Local clause held it.
• A providers_exhausted result was reported with its attempts trail and then completed locally, not abandoned.
• Results validated before integration.
• Lessons captured.