/security-flow-readiness

name: security-flow-readiness
description: "Phase 1 Readiness of security-flow"
disable-model-invocation: true
user-invocable: false
alwaysApply: false
tags: ["security", "workflow-phase"]
baseSchema: docs/schemas/phase.md

<security_flow_readiness>

<description_and_purpose> Establish minimal context and gate secret-bearing files before model source ingestion. </description_and_purpose>

<workflow_context> Phase 1. Mandatory `executor`; bounded metadata and commands only. </workflow_context>

<phase_steps> 1. Inventory request metadata 2. Inventory available tools 3. Run filename-only secret gate 4. Return gate decision </phase_steps>

<execute_readiness step="1.1" subagent="executor" role="Bounded readiness and secret-gate operator" subagent_required_model="Claude Haiku 4.5, GPT-5.6 Terra, Gemini 3.7 Flash, GPT-5.6 Luna">

1. USE SKILL `subagent-directives`. 2. USE SKILL `security`. 3. USE SKILL `sensitive-data`. 4. Read request metadata only: target names, paths, environment labels, requested review type. 5. List installed/reachable tools without installation, authentication, or network calls. 6. Prefer an approved filename-only secret scanner. 7. Otherwise USE SKILL `security` for its filename-only fallback. 8. Return affected filenames only, never matches or content. 9. Classify environment from approved metadata, not file values. 10. Update `security-flow-state.md`.

</execute_readiness>

<gate step="1.2">

Return exactly one token:

• No candidate files: PASS.
• DEV/QA envs only: NEEDS-HITL; recommend exclusions.
• Above-QA or ambiguous: STOP-HIGH-RISK.
• Scanner unusable or scanner exit 2: STOP-SCANNER-UNUSABLE; do not ingest source.

</gate>

<validation_checklist>

• No source content entered agent context.
• Output contains filenames and metadata only.
• Tool inventory records availability, not assumptions.
• Gate decision matches the strictest environment.

</validation_checklist>

</security_flow_readiness>