compilelisted
Install: claude install-skill trevhud/rote
# Compile a skill
You orchestrate the `rote` CLI. It runs an LLM compiler agent over a
source skill and emits a deterministic pipeline. Your job: resolve the
inputs, run the CLI, then interpret the output for the user. You never
classify nodes or write pipeline.yaml yourself — the CLI's agent does.
## 1. Identify the source skill
The source is a **directory containing a `SKILL.md`** (optionally a
`references/` folder). The user names it, or you infer it from context
(a skill just discussed, a path in the conversation, `.claude/skills/*`
or `skills/*` in the project).
**Confirm the resolved absolute path with the user before running.**
Compilation costs real time and tokens; never guess-and-go. If the
directory has no `SKILL.md`, stop and ask.
<!-- rote:worth -->
### Is it worth compiling?
Compiling to the skill runtime pays off when most of the skill's work is
fixed steps: tool calls, lookups, filtering, arithmetic, formatting. The
compiled skill does those in code and stops for you only where the skill
needs judgment. Each stop is a round trip that re-reads the conversation,
so a skill that is mostly judgment (search, read, then write an answer)
runs slower and costs more compiled. Measured: Atlassian's
search-company-knowledge took 4x longer at 3x the cost, while
expense-audit ran 3.6x faster at 62% less.
Read the SKILL.md before compiling, and count the steps that need
judgment: choosing, ranking, summarizing, writing prose. If a run needs
more than two, tell the us