manager

Sub-agent dispatch reference

Source: metyatech/skill-sub-agent-dispatch. To update this skill, edit the repository and push. The agent MUST NOT edit the installed copy.

This skill contains reference data and detailed procedures for dispatching sub-agents. The always-loaded invariants for sub-agent dispatch (agents-mcp tooling, ai-quota, mode, decision framework, verification deliverable format, cost discipline, execution discipline, lifecycle) live in the sub-agent-dispatch rule module. Spawning prompt requirements and delegated-agent obligations live in the delegation rule module. Approval rules live in the approval-gates rule module.

Dispatch workflow

1. Receive the user's request.
2. Decompose into discrete work items. Track them.
3. Classify each item using the Decision framework defined in the sub-agent-dispatch rule.
4. Dispatch all independent items in parallel.
5. Report to the user: what was dispatched, what is pending. Return control immediately.
6. On every subsequent user message, the agent MUST call Status(wait=false) first and report any state changes before addressing the user's new request.
7. When all sub-agents are done and review gates pass, the agent MUST summarize results and proceed to next steps.

Cross-agent invocation (agents-mcp)

agents-mcp is the metyatech MCP server that works uniformly from Claude Code, Codex, Gemini CLI, and other agent CLIs.

One-time setup

See README.md for platform-specific installation commands.

Dispatching a task

Use agents-mcp to spawn sub-agents with these parameters:

• prompt — the full self-contained task description.
• agent_type — target agent (claude, codex, copilot). The agent MUST NOT use gemini (see Gemini section below).
• model — explicit model string from the Model Inventory (e.g., "claude-sonnet-4-6").
• effort — optional reasoning effort. Set from the Model Inventory; omit when not needed.
• mode: 'edit' — REQUIRED for implementation agents.

Monitoring

After spawning agents, return to the user immediately. Use background or non-blocking monitoring to be notified when agents finish. agents-mcp wait and Status(wait=true) both poll until completion or timeout.

If agents-mcp is not configured

Stop and report the limitation to the user. The agent MUST NOT simulate or substitute the work itself.

Verification prompt templates

Implementation-agent template

Delegated mode. Implement approved scope only.
Return exactly: (1) Restated AC, (2) AC->evidence with exact commands/manual steps + PASS/FAIL/NOT RUN, (3) Files changed, (4) Assumptions/uncertainties, (5) Risks/rollback notes.
If spec/behavior text changed: cite file + section and summarize behavior delta.
Run applicable tests/verification and include exact commands and outputs summary.

Review-agent template

Delegated mode reviewer. Do not implement; review only.
Original requirements: [paste original AC/spec here]
Implementation output: [paste implementation agent's report here]
Note: tests were written by an agent and may confirm implementation behavior rather than spec intent.
Return: explicit PASS or FAIL, reasons, spec alignment (does the implementation meet the original requirements, not just the tests?), AC coverage gaps, and any ambiguous points.
Reject if evidence format is incomplete, outcomes are unsupported, or implementation deviates from original spec.

Spec-change protocol

• If behavior or spec text changes, the implementation agent MUST cite where it changed (file + section heading) and MUST summarize the behavior delta.
• The reviewer MUST explicitly confirm spec alignment and list any ambiguous points requiring follow-up.

Reviewer model selection

• For Standard tier reviews when verification fails: use claude-haiku-4-5-20251001.
• For Heavy tier reviews and release/production changes: use claude-sonnet-4-6 with medium effort.

Error handling

• If a sub-agent fails, call Status and treat agents[].errors and agents[].diagnostics.tail_errors as the primary failure reason.
• Open raw logs only when needed via agents[].diagnostics.log_paths.stdout (e.g., tail the file). Do not manually hunt logs.

Model inventory and routing

Last reviewed: 2026-02-27. Update this section when models change.

Tier definitions

• Free — trivial lookups, simple Q&A, straightforward single-file edits. Copilot only.
• Light — mechanical transforms, formatting, simple implementations, quick clarifications.
• Standard — general implementation, code review, multi-file changes, most development work.
• Heavy — architecture decisions, safety-critical code, complex multi-step reasoning.
• Large Context — tasks requiring more than 200k token input.
• Bulk/Idle — automated compliance checks, repetitive transforms across repos. Prefer lowest-cost agents.

Classify each task into a tier, then pick an agent with available quota and select the ★ preferred model for that tier. Fall back to other models in the same tier when the preferred model's agent has no quota.

Claude

Tier	Model	Effort	Notes
Light	claude-haiku-4-5-20251001	—	Effort not supported; SWE-bench 73%
Standard	claude-sonnet-4-6	medium	★ Default; SWE-bench 80%
Heavy	claude-opus-4-6	high	SWE-bench 81%; max effort for hardest tasks

Codex

Tier	Model	Effort	Notes
Light	gpt-5.1-codex-mini	medium	medium/high only
Standard	gpt-5.3-codex	medium	★ Latest flagship; SWE-bench Pro 57%
Standard	gpt-5.2-codex	medium	Previous gen; SWE-bench Pro 56%
Standard	gpt-5.2	medium	General-purpose; best non-codex reasoning; SWE-bench 80%
Heavy	gpt-5.3-codex	xhigh	★ Best codex at max effort
Heavy	gpt-5.1-codex-max	xhigh	Extended reasoning; context compaction
Heavy	gpt-5.2-codex	xhigh	Alternative
Heavy	gpt-5.2	xhigh	General reasoning fallback

Gemini sub-agent reliability

⚠ The gemini agent type MUST NOT be used for sub-agent delegation. Gemini CLI agents fail with HTTP 429 "No capacity available" server errors too frequently to be reliable, even for single unattended tasks. Gemini CLI MAY be used interactively when invoked directly by the user. For large-context work, use Copilot with the gemini-3-pro model instead.

Tier	Model	Effort	Notes
Light	gemini-3-flash-preview	—	SWE-bench 78%
Standard	gemini-3-pro-preview	—	★ 1M token context; SWE-bench 76% (interactive only)
Large Context	gemini-3-pro-preview	—	>200k token tasks; 1M context (interactive only)

Aider (BYOK)

Aider uses external API keys. Cost is per-token from the provider (no subscription). Best for bulk/idle tasks with cheap API models.

Tier	Model	Provider	Notes
Bulk/Idle	deepseek/deepseek-chat	DeepSeek	★ $0.07-$0.27/MTok input; SWE-bench 73%
Bulk/Idle	deepseek/deepseek-reasoner	DeepSeek	$1.30/Aider task
Light	claude-3-5-haiku-20241022	Anthropic	Same as Claude Haiku
Standard	claude-sonnet-4-6-20261022	Anthropic	Same as Claude Sonnet

Configure via DEEPSEEK_API_KEY or ANTHROPIC_API_KEY. Aider is not yet integrated into agent-runner (tracked task).

Copilot

Copilot charges different quota per model. Prefer lower-multiplier models when task complexity allows. Effort is not configurable (ignored).

Tier	Model	Quota	Notes
Free	gpt-5-mini	0x	★ SWE-bench ~70%
Free	gpt-4.1	0x	1M context; SWE-bench 55%
Light	claude-haiku-4-5	0.33x	★ SWE-bench 73%
Light	gpt-5.1-codex-mini	0.33x	Mechanical transforms
Standard	claude-sonnet-4-6	1x	★ Default; SWE-bench 80%
Standard	gpt-5.3-codex	1x	Latest codex flagship
Standard	gpt-5.2	1x	Best general reasoning
Standard	gpt-5.2-codex	1x	Agentic coding
Standard	gpt-5.1-codex-max	1x	Extended reasoning
Standard	claude-sonnet-4-5	1x	SWE-bench 77%; prefer 4.6
Standard	gpt-5.1-codex	1x	SWE-bench 77%
Standard	gpt-5.1	1x	General purpose; SWE-bench ~76%
Standard	gemini-3-pro	1x	1M context; SWE-bench 76%
Standard	claude-sonnet-4	1x	Legacy; last choice
Heavy	claude-opus-4-6	3x	★ SWE-bench 81%
Heavy	claude-opus-4-5	3x	SWE-bench 81%; prefer 4.6
—	claude-opus-4-6 fast	30x	Avoid; excessive quota cost

Routing principles

• Active sub-agent types are claude, codex, copilot. The gemini agent type MUST NOT be spawned (see Gemini section above). All active agent types operate on independent flat-rate subscriptions. Route by model quality, quota conservation, and quota distribution.
• All agents can execute code, modify files, and perform multi-step tasks. Route by model quality and quota, not by execution capability.
• Spread work across agents to maximize total throughput.
• For large-context tasks (>200k tokens), prefer Copilot with the gemini-3-pro model. Do NOT use the gemini agent type.
• For trivial tasks, prefer Copilot free-tier models (0x quota) before consuming other agents' quota.
• When multiple agents can handle a task equally well, prefer the one with the most remaining quota.

Cost-efficiency priority

• Prefer flat-rate (subscription) agents over pay-per-token agents when quota is available.
• Idle/bulk tasks MUST NOT consume premium quota. Route to: Copilot 0x → Aider+DeepSeek → Amazon Q.
• Reserve premium quota for interactive use. Usage gates enforce minimum 70% remaining for Claude and Codex 5-hour windows.
• Sonnet 4.6 is the workhorse. Default to Sonnet; escalate to Opus for Heavy tier or when strict rule compliance failures occur. Use medium effort for both — research shows higher effort degrades instruction-following on multi-constraint rule sets (arXiv:2505.11423).

Quota fallback logic

If the primary agent has no remaining quota:

1. Query quota for all agents via ai-quota.
2. Select any agent with available quota that has a model at the required tier.
3. For Copilot fallback, prefer lower-multiplier models to conserve quota.
4. For Standard tier with no subscription quota: fall back to Aider + DeepSeek (pay-per-token, ~$1/task).
5. If the fallback model is significantly less capable, note the degradation in the dispatch report.
6. If no agent has quota and no pay-per-token fallback is available, queue the task and report the block immediately. Do not drop silently.

Routing decision sequence

1. Classify the task tier (Free / Light / Standard / Heavy / Large Context / Bulk/Idle).
2. For Free tier: dispatch to Copilot with a 0x model. Skip quota check.
3. For Bulk/Idle tier: dispatch to Copilot 0x → Aider+DeepSeek → Amazon Q (in that order). Skip premium agents and do not use the Gemini agent type.
4. For other tiers: check quota for all agents via ai-quota.
5. Pick the agent with available quota at the required tier; prefer the agent with the most remaining quota when multiple qualify.
6. Set agent_type, model, and effort from the tables above (omit effort when column shows —).
7. If primary choice has no quota: apply fallback logic.
8. Include the chosen agent, model, tier, and effort in the dispatch report.

Notable models (not in current routing)

These models are not yet accessible via supported agent CLIs but are worth monitoring:

Model	Provider	Strength	Access
Kimi K2.5	Moonshot AI	IFEval SOTA 94.0	API only; no CLI
Grok 4.20	xAI	4-agent council; 2M context	xAI API only
Qwen 3.5	Alibaba	IFEval 92.6	API only
GLM-4.7	Zhipu AI	Dual-judge highest score	API only

Platform-specific workarounds

• On platforms with policy-blocked file deletion commands (e.g., Codex on Windows), see the Codex-specific workarounds in the command-execution rule module.