skills.rest
code-yeongyu
code-yeongyuCOMMUNITY
Explore all skills by this author →
Website𝕏@q_yeon_gyu_kimGitHub

senpi-qa

Validate the Senpi adapter and task engine against live binary drivers in isolated environments.

68.6k|5.6k|Updated Dec 3, 2025
Search Tags:#QA#Testing#Automation#adapter#senpi#Sandbox
One-click install
npx skills add https://github.com/code-yeongyu/oh-my-opencode --skill senpi-qa
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: senpi-qa
Source: https://github.com/code-yeongyu/oh-my-opencode/tree/main/.agents/skills/senpi-qa
Command: npx skills add https://github.com/code-yeongyu/oh-my-opencode --skill senpi-qa

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill addresses the difficulty of verifying complex AI agent adapters and task engines without polluting the developer's local environment or relying on unreliable unit tests.

Core Features & Use Cases

  • Isolated QA Harness: Executes the real senpi binary within a sandboxed SENPI_CODING_AGENT_DIR to ensure zero impact on the user's actual agent configuration.
  • Canonical Evidence Management: Enforces a strict, predictable path structure for all artifacts, ensuring that QA results are always discoverable and citable in pull requests.
  • Use Case: When modifying the packages/omo-senpi adapter or task lifecycle logic, use this Skill to run end-to-end tests that verify task streaming, RPC contracts, and team delivery without risking local state corruption.

Quick Start

Use the senpi-qa skill to run the adapter driver and generate evidence for the current branch.

Frequently Asked Questions about senpi-qa

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a live QA test for the Senpi adapter?▼

Use the provided driver scripts in packages/omo-senpi/scripts/qa/ to execute the senpi binary. Ensure you resolve a unique evidence directory using the resolve-evidence-dir.mjs script before running the driver to capture all output artifacts.

Does this Skill modify my local ~/.senpi/agent configuration?▼

No, the live drivers are designed to ignore the caller's agent directory. They create and use an isolated SENPI_CODING_AGENT_DIR for every run, ensuring your real agent configuration remains untouched.

What should I do if the senpi binary is missing?▼

The live drivers will report a SKIP or FAIL status if the binary is not found. Do not attempt to bypass this; record the skip in the evidence README to document that the QA run could not be completed.

Why is the evidence directory path strictly enforced?▼

Strict path enforcement ensures that all QA artifacts are stored in a predictable location (.omo/evidence/omo-senpi-adapter/<slug>/). This allows reviewers to easily locate and verify the results of your tests without searching through local-ignore directories.

Related Skills

Explore More Software Engineering Skills→
basilisk-labsbasilisk-labs
72

agentplane

Plan, execute, and verify AI agent tasks with git-native evidence.

Official
Advanced
daodaoedudaodaoedu

openspec-archive-change

Archive completed OpenSpec change directories after validating artifacts and tasks.

Official
Advanced
Ninesun9Ninesun9
3

skill demo

Provide a minimal skill unit for testing skill system parsing and activation.

Community
Basic
interserverinterserver
1

mocha-skill

Generate Jest-free unit tests with Mocha, Chai, and Sinon.

Official
Intermediate