active/showboat-v2/SKILL.md
Create schema-backed Showboat integration proofs with scenario summaries, raw artifacts, and replayable verification.
npx skillsauth add kevinslin/skills showboat-v2Install this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Use this skill when the user wants an executable Showboat proof that also has a durable, structured proof directory. It is for live integration or behavior proof, not for wrapping an existing unit test run as the whole proof.
Use ../schemas/SKILL.md to inspect and materialize the directory structure before writing proof content. Prefer the bundled integ-proof schema when the user does not name another schema.
Before writing a durable proof directory or any long-lived proof artifact, invoke ../mem/SKILL.md to resolve the intended knowledge base, root, schemas, and file rules. Do this because the memory root may be any configured path, not necessarily a directory named .mem. Only write outside $mem when the artifact is explicitly temporary, repo-owned source/docs, or the user names a concrete non-memory destination.
$mem before choosing <proofs-root> and record the selected base name, resolved root, and concrete proof path. Do not assume a specific repository, connector, plugin, app, or artifact root.../schemas/scripts/schema.py show integ-proof
../schemas/scripts/schema.py materialize integ-proof \
--out <proofs-root> \
--var proof=<proof-slug> \
--var scenario=<scenario-slug> \
--include <proof-slug>/proof \
--include <proof-slug>/scenario/<scenario-slug> \
--skip-existing
<proof-root> as <proofs-root>/<proof-slug>, then create the dynamic artifact directories before writing into them:mkdir -p <proof-root>/raw <proof-root>/scripts
Keep bulky logs, transcripts, screenshots, JSON, and nondeterministic command output in raw/; keep proof-local collection or summarization helpers in scripts/.
6. Confirm Showboat is available:
uvx showboat --help
raw/, usually raw/showboat-summary.md:uvx showboat init <proof-root>/raw/showboat-summary.md "<Proof Title>"
uvx showboat note <proof-root>/raw/showboat-summary.md "<short context>"
exec command that summarizes or validates the captured raw artifact with deterministic output, so the verified Showboat document is tied to the observed live behavior:uvx showboat exec <proof-root>/raw/showboat-summary.md bash "<deterministic summary-or-validation command>"
uvx showboat verify <proof-root>/raw/showboat-summary.md
scenario/<scenario>.md.proof.md last with the claim, expected behavior, target, status, scenario result table or bullets, relevant scripts, and raw artifact index.Use the materialized schema template for each scenario file. Keep it concise and fill these sections:
# Scenario: {title}
## Purpose
{one line}
## Preconditions
## Config
## Action
## Expected
## Observed
## Related
- raw artifacts:
## Notes
Omit or mark Config as not applicable only when the scenario genuinely has no relevant configuration. Link proof-local summarizers under Related or Notes when they are relevant.
schemas; do not invent a parallel layout when a schema exists.Observed, mark the proof status accordingly, and keep the raw artifact that explains the failure.For the default integ-proof schema, the finished proof should look like this:
<proof-slug>/
├── proof.md
├── scenario/
│ └── <scenario-slug>.md
├── scripts/
└── raw/
Use additional files or nested folders under raw/ and scripts/ when the proof needs them.
development
Resolve explicit shortcut triggers and usage. Always read this file at the start of a thread or when user mentions `trigger`.
testing
Improve skills from observed agent friction in sessions, PRs, or audits.
development
Generate incremental Slack digests for channels, topics, and categories.
testing
Audit an OpenClaw maturity-scorecard surface into an evidence-backed component score report. Use when given a surface from an OpenClaw maturity-scorecard.md and asked to score coverage, quality, readiness, or generate a detailed surface report plus per-component subreports.