원클릭으로 Manus에서 모든 스킬 실행

skill-tester

스타0

포크0

업데이트2026년 2월 19일 15:01

Run local tests for each OpenClaw skill before publishing to ClawHub. Discovers skills, runs configured or heuristic tests, reports pass/fail. Use to ensure all skills work as intended before upload.

설치

Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.

Manus에서 실행

출처

austindixson

austindixson/overclaw

GitHub 저장소 열기 Creator 저장소 보기

다운로드

Manus에서 실행

Skill Tester

Description

Run local tests for each OpenClaw skill before publishing to ClawHub. Discovers skills, runs configured or heuristic tests, reports pass/fail. Use to ensure all skills work as intended before upload.

Skill Tester

Run robust local tests on your OpenClaw skills before uploading to ClawHub. Discovers skills in workspace/skills, runs test commands (from skill_tests.json or heuristics), and outputs a pass/fail report.

Usage

Before publishing skills to ClawHub: run --all to verify every skill.
After changing a skill: run --skill SLUG to test that one.
In CI or as a sub-agent task: use --json for machine-readable report.

python3 <skill-dir>/scripts/skill_tester.py --list
python3 <skill-dir>/scripts/skill_tester.py --all [--json] [-v]
python3 <skill-dir>/scripts/skill_tester.py --skill <slug> [--json] [-v]

--list — List discoverable skills (have SKILL.md or _meta.json). No tests run.
--all — Run tests for every discovered skill.
--skill SLUG — Run tests only for the given skill (e.g. agent-swarm, gateway-guard).
--json — Output a JSON report: { "skills": { ... }, "passed": N, "failed": M }. Exit 1 if any failed.
-v / --verbose — Print each test name and result per skill.

Test config

Tests are defined in scripts/skill_tests.json (next to skill_tester.py). Each key is a skill slug; value is a list of test objects:

name — Test label in output.
cmd — List of command + args (e.g. ["python3", "scripts/router.py", "spawn", "--json", "hello"]).
cwd — Optional subdir of the skill (default: skill root).
expect_exit — Expected process exit code (default 0).
expect_json_keys — Optional list of keys that must exist in stdout JSON.
timeout — Seconds (default 25).

If a skill has no entry in skill_tests.json, the tester runs a heuristic: for each scripts/*.py, try --help or --json and treat success as pass.

Requirements

OPENCLAW_HOME set (or ~/.openclaw) with workspace/skills/ containing skill dirs.
Python 3.6+.
Skills under test must have their own dependencies (e.g. gateway-guard needs openclaw CLI for full ensure; status --json can still be tested).

Exit codes

0 — All tested skills passed.
1 — At least one test failed.
2 — Usage error (e.g. missing --skill/--all, or skills dir not found).

이 저장소의 다른 Skills

같은 저장소

nanobot-overstory-bridge

austindixson/overclaw

Seamless bidirectional bridge between nanobot (Ollama Mistral orchestrator) and overstory (Claude Code agent swarm). Routes tasks through the OverClaw gateway (port 18800) to overstory for subagent coordination, syncs memory.

2026-02-210

funclip-commander

austindixson/overclaw

Integrates OpenClaw with the FunClip AI video editor to enable AI-powered video recognition and clipping via command line.

2026-02-190

agent-swarm

austindixson/overclaw

IMPORTANT: OpenRouter is required. Routes tasks to the right model and always delegates work through sessions_spawn. Rejects prompt-injection patterns in task strings (v1.7.6+).

2026-02-190

auto-clipper

austindixson/overclaw

Automatically create clips and videos from media files in a specified folder. Uses Agent Swarm for intelligent task delegation and supports cron-based scheduling.

2026-02-190

blog-generator

austindixson/overclaw

Analyzes journal entries and chat history to identify high-value topics and automatically generate blog posts.

2026-02-190

chat-history-analyzer

austindixson/overclaw

Extracts and analyzes Cursor IDE chat history to identify key discoveries, obstacles, and solutions, saving findings to the journal.

2026-02-190

name	skill-tester
description	Run local tests for each OpenClaw skill before publishing to ClawHub. Discovers skills, runs configured or heuristic tests, reports pass/fail. Use to ensure all skills work as intended before upload.
version	1.0.0

Skill Tester

Description

Run local tests for each OpenClaw skill before publishing to ClawHub. Discovers skills, runs configured or heuristic tests, reports pass/fail. Use to ensure all skills work as intended before upload.

Skill Tester

Usage

Before publishing skills to ClawHub: run --all to verify every skill.
After changing a skill: run --skill SLUG to test that one.
In CI or as a sub-agent task: use --json for machine-readable report.

python3 <skill-dir>/scripts/skill_tester.py --list
python3 <skill-dir>/scripts/skill_tester.py --all [--json] [-v]
python3 <skill-dir>/scripts/skill_tester.py --skill <slug> [--json] [-v]

--list — List discoverable skills (have SKILL.md or _meta.json). No tests run.
--all — Run tests for every discovered skill.
--skill SLUG — Run tests only for the given skill (e.g. agent-swarm, gateway-guard).
--json — Output a JSON report: { "skills": { ... }, "passed": N, "failed": M }. Exit 1 if any failed.
-v / --verbose — Print each test name and result per skill.

Test config

Tests are defined in scripts/skill_tests.json (next to skill_tester.py). Each key is a skill slug; value is a list of test objects:

name — Test label in output.
cmd — List of command + args (e.g. ["python3", "scripts/router.py", "spawn", "--json", "hello"]).
cwd — Optional subdir of the skill (default: skill root).
expect_exit — Expected process exit code (default 0).
expect_json_keys — Optional list of keys that must exist in stdout JSON.
timeout — Seconds (default 25).

If a skill has no entry in skill_tests.json, the tester runs a heuristic: for each scripts/*.py, try --help or --json and treat success as pass.

Requirements

OPENCLAW_HOME set (or ~/.openclaw) with workspace/skills/ containing skill dirs.
Python 3.6+.
Skills under test must have their own dependencies (e.g. gateway-guard needs openclaw CLI for full ensure; status --json can still be tested).

Exit codes

0 — All tested skills passed.
1 — At least one test failed.
2 — Usage error (e.g. missing --skill/--all, or skills dir not found).