git:20260820.bf0eddd to git:20260916.17858cf

2 added, 1 removed. Audit A to A.

---

name: loop-research

- description: Author and optionally direct-launch a two-node loop-graph run that compares feasible technical approaches with open-source evidence, primary research, and controlled experiments before selecting one. Use when a decision needs several research and evaluation rounds with durable evidence and independent audit. Not for implementing an already chosen requirement, code cleanup, or a quick literature summary. Do not execute or resume generated runtime node files from this authoring skill.

+ description: Use only when the user explicitly invokes /loop-research or names this skill. Authors a loop-research run. Do not use unless named. Do not execute or resume generated runtime node files.

+ disable-model-invocation: true

---

# loop-research — a loop-graph preset for evidence-led choices

A thin authoring entry. It binds a research-and-selection pack, starts the owner

interview, then follows [`loop-graph`](../loop-graph/SKILL.md) to compile the normal

executor, ledger, directives, ops, and supervisor artifacts. It is not a research node,

a second runtime, or a second template set.

## Fit check

- Use this when the approach is undecided and a decision needs comparative evidence from

open-source projects, primary research, and a controlled benchmark or A/B experiment.

- Use [`../loop-deliver/SKILL.md`](../loop-deliver/SKILL.md) once the approach is chosen,

or [`../loop-converge/SKILL.md`](../loop-converge/SKILL.md) for code consolidation.

- For a short answer, one source lookup, or non-comparative literature summary, use the

host's ordinary research task instead of a graph.

## On invoke

1. Inspect the workspace, existing evidence, experiment harnesses, data policy, and

current host the same way loop-graph does. Never ask which client this is when context

already identifies it.

2. Read and bind [`preset.md`](preset.md). That pack is the North Star, supervisor

requirement, interview, evidence shape, method guards, knob overrides, and artifact

emphasis. Do not redesign them.

3. Start the owner interview immediately. Ask only the pack's unresolved choices —

decision/scope, evidence budget and data authority, and launch — as recommended A/B

(or A/B/C) choices. Do not ask the owner to invent evaluation criteria.

4. Read and follow [`../loop-graph/SKILL.md`](../loop-graph/SKILL.md) from **When called

from a preset skill** through generate and deliver. Compile only from

loop-graph's [`templates/`](../loop-graph/templates/). This skill never executes the

generated nodes.