Add SecPriva security/privacy agent skill definitions and OpenAI configs

[SaltProphet]

Motivation

• Introduce a set of reusable SecPriva agent skills to standardize security and privacy reviews across the repo.
• Capture workflows and expected outputs for common tasks like API security, SAST, PII discovery, retention/deletion, incident triage, threat modeling, CI gates, and audit evidence.
• Provide per-agent OpenAI configuration to drive consistent assistant behavior using the gpt-5 model.

Description

• Added new skill definition files at /.github/skills/<secpriva-*>/SKILL.md for the following agents: secpriva-api-security-agent, secpriva-audit-evidence-agent, secpriva-ci-security-gate-agent, secpriva-incident-triage-agent, secpriva-pii-discovery-agent, secpriva-retention-deletion-agent, secpriva-sast-agent, and secpriva-threat-modeling-agent.
• Added corresponding agent configs at /.github/skills/<secpriva-*>/agents/openai.yaml that set name, model: gpt-5, and purpose-specific instructions for each agent.
• Each SKILL.md includes a short description, recommended use cases, a stepwise workflow, and an expected output format to standardize responses.

Testing

• No automated tests were run for these documentation/config additions.

---

Codex Task

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
secpriva	Ready	Preview, Comment	Mar 5, 2026 7:48pm