CS

coveros/strategiesfortestingai

Productivity workflow

このページは AI により自動翻訳されています。技術的な詳細は原文をご確認ください。 原文を見る

14 stars 品質 40 トレンド 28

Repo for Strategies for Testing AI-Based Systems Class

概要

This folder contains reusable GitHub Copilot customization assets for a 1-hour AppSec talk. - copilot-instructions.md: Workspace-level defaults for AppSec demo behavior. - instructions/appsec-demo.instructions.md: Task and file-scoped instructions for security analysis and remediation suggestions. - agents/appsec-triage.agent.md: Custom agent focused on evidence-based vulnerability triage. - agents/remediation-coach.agent.md: Custom agent focused on minimal, reviewable patch suggestions. - skills/vulnerability-discovery/SKILL.md: Structured discovery workflow. - skills/secure-patch-suggestion/SKILL.md: Patch suggestion workflow. - skills/remediation-validation/SKILL.md: Post-fix validation workflow. - skills/runtime-remediation/SKILL.md: Controlled live code-edit workflow for demo apply mode. 1. Run analysis with AppSec Triage Agent. 2. Produce patch suggestions with Remediation Coach Agent. 3. Optionally switch to apply mode for runtime edits with runtime-remediation. 4.

README

Custom Agent Assets for AppSec Demo

This folder contains reusable GitHub Copilot customization assets for a 1-hour AppSec talk.

What is included

  • copilot-instructions.md: Workspace-level defaults for AppSec demo behavior.
  • instructions/appsec-demo.instructions.md: Task and file-scoped instructions for security analysis and remediation suggestions.
  • agents/appsec-triage.agent.md: Custom agent focused on evidence-based vulnerability triage.
  • agents/remediation-coach.agent.md: Custom agent focused on minimal, reviewable patch suggestions.
  • skills/vulnerability-discovery/SKILL.md: Structured discovery workflow.
  • skills/secure-patch-suggestion/SKILL.md: Patch suggestion workflow.
  • skills/remediation-validation/SKILL.md: Post-fix validation workflow.
  • skills/runtime-remediation/SKILL.md: Controlled live code-edit workflow for demo apply mode.

Intended use

  1. Run analysis with AppSec Triage Agent.
  2. Produce patch suggestions with Remediation Coach Agent.
  3. Optionally switch to apply mode for runtime edits with runtime-remediation.
  4. Re-check findings against the baseline standard in docs/appsec-demo/appsec-baseline-standard.md.

Best practice for demos:

  • Explicitly select the target custom agent before each prompt.
  • Explicitly invoke skills (/vulnerability-discovery, /secure-patch-suggestion, /runtime-remediation, /remediation-validation) instead of relying on implicit routing.

Scripted option for live runtime edits:

  • python appsec_runtime_remediation_demo.py --mode assess
  • python appsec_runtime_remediation_demo.py --mode apply
  • python appsec_runtime_remediation_demo.py --mode rollback

Speaker support:

  • docs/appsec-demo/appsec-speaker-cheat-sheet.md
  • docs/appsec-demo/appsec-live-demo-script.md
  • docs/appsec-demo/appsec-live-demo-one-page.md

Notes

  • These assets are demo-oriented and intentionally lightweight.
  • Default behavior is suggestion-only remediation.
  • Apply mode is supported when explicitly requested, with reversible-edit and rollback guidance.
GitHub で README 全文を見る

推奨ツール

別のキーワードを試すか、フィルタを外してください。

インストール

npx skillfish add coveros/strategiesfortestingai
MCP 初心者?mcpskill.org のガイドから始めましょう →