GravionLabs/elysion / .agents/skills/deliver
deliver
VerifiedDeliver a PBI, Bug, Feature or Epic item by item with subagents (implement, review, PR, CI, merge). Use for "/deliver 12" or "deliver feature 7".
references/roles.md
View on GitHub# Roles: model and effort
The skill gives its subagents `model` and `effort` of the roles `implement`, `review`, `fix` and `release-review` (step 9, the review of the whole release diff, `references/release-review.md`) when the harness takes a model per call (Claude Code); OpenCode and Copilot CLI use the session model or their own agent config. The other roles of the table are kept in the config schema; this skill does not use them. A run of many PBIs on the session model is expensive (about 30 k context tokens per subagent before it starts), so each role has a default.
## Defaults
| Role | Model | Effort |
| :------------------------------------- | :----------------- | :------------- |
| `select`, `create`, `gate`, `pr`, `ci` | haiku | low |
| `review` | sonnet | medium |
| `release-review` | opus | high |
| `plan` | session model | session effort |
| `implement`, `fix` | by tier of the PBI | by tier |
Tier (`Tier:` line of the PBI body; else the size: S fast, M standard, L strong; else standard): `fast` haiku/low, `standard` sonnet/medium, `strong` opus/high.
The tier is shown before and during a run: the select line (`#<nr> select: next (tier standard → sonnet/medium)`; `tier from size M` or `default` when the body has no `Tier:` line), the tier distribution of the open items in the report, and the loop's `--model` per PBI (`references/loop.md`). The loop maps the tier to the model: `fast` haiku, `standard` sonnet, `strong` opus; `-Model` overrides it for all PBIs.
## Override
`deliver.roles` in `.agents/config.json` (schema `schema/config.schema.json` of the `implementation` skill). A role entry replaces `model` and `effort` of that role; `tiers` (only `implement` and `fix`) replaces single tiers. Unset keys keep the default. `plan` has no default, so it uses the session model until configured.
```json
{
"deliver": {
"roles": {
"review": { "model": "opus", "effort": "high" },
"implement": { "tiers": { "standard": { "model": "opus", "effort": "medium" } } },
"gate": { "model": "sonnet" }
}
}
}
```
`model` is what the harness accepts per subagent (`sonnet`, `opus`, `haiku`, `fable`), `effort` one of `low`, `medium`, `high`, `xhigh`, `max`.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |