sagaz-ai 0.1.0

package/ai-orchestration-ecosystem/protocols/github-operations.md

@@ -0,0 +1,32 @@
+# Protocol: Github Operations
+
+## Objective
+
+Define how Sagaz should handle Github Operations while keeping the process clear, low-token, safe, and verifiable.
+
+## Required Practice
+
+- Start from the user goal and current project state.
+- Load only relevant context.
+- Separate facts, assumptions, inferences, risks, and decisions.
+- Ask permission before meaningful state changes.
+- Record evidence and residual risk.
+
+## Standard Recommendation Format
+
+```md
+Recommendation:
+Why now:
+What changes:
+Benefit:
+Risk:
+Permission required:
+```n
+## Blocking Conditions
+
+- The primary flow fails.
+- A relevant build, check, or test fails without explanation.
+- Secrets or sensitive data would be exposed.
+- A high risk is not accepted by the user.
+- Verification evidence is missing for the risk level.
+

package/ai-orchestration-ecosystem/protocols/guided-proactivity.md

@@ -0,0 +1,32 @@
+# Protocol: Guided Proactivity
+
+## Objective
+
+Define how Sagaz should handle Guided Proactivity while keeping the process clear, low-token, safe, and verifiable.
+
+## Required Practice
+
+- Start from the user goal and current project state.
+- Load only relevant context.
+- Separate facts, assumptions, inferences, risks, and decisions.
+- Ask permission before meaningful state changes.
+- Record evidence and residual risk.
+
+## Standard Recommendation Format
+
+```md
+Recommendation:
+Why now:
+What changes:
+Benefit:
+Risk:
+Permission required:
+```n
+## Blocking Conditions
+
+- The primary flow fails.
+- A relevant build, check, or test fails without explanation.
+- Secrets or sensitive data would be exposed.
+- A high risk is not accepted by the user.
+- Verification evidence is missing for the risk level.
+

package/ai-orchestration-ecosystem/protocols/memory.md

@@ -0,0 +1,32 @@
+# Protocol: Memory
+
+## Objective
+
+Define how Sagaz should handle Memory while keeping the process clear, low-token, safe, and verifiable.
+
+## Required Practice
+
+- Start from the user goal and current project state.
+- Load only relevant context.
+- Separate facts, assumptions, inferences, risks, and decisions.
+- Ask permission before meaningful state changes.
+- Record evidence and residual risk.
+
+## Standard Recommendation Format
+
+```md
+Recommendation:
+Why now:
+What changes:
+Benefit:
+Risk:
+Permission required:
+```n
+## Blocking Conditions
+
+- The primary flow fails.
+- A relevant build, check, or test fails without explanation.
+- Secrets or sensitive data would be exposed.
+- A high risk is not accepted by the user.
+- Verification evidence is missing for the risk level.
+

package/ai-orchestration-ecosystem/protocols/model-routing.md

@@ -0,0 +1,32 @@
+# Protocol: Model Routing
+
+## Objective
+
+Define how Sagaz should handle Model Routing while keeping the process clear, low-token, safe, and verifiable.
+
+## Required Practice
+
+- Start from the user goal and current project state.
+- Load only relevant context.
+- Separate facts, assumptions, inferences, risks, and decisions.
+- Ask permission before meaningful state changes.
+- Record evidence and residual risk.
+
+## Standard Recommendation Format
+
+```md
+Recommendation:
+Why now:
+What changes:
+Benefit:
+Risk:
+Permission required:
+```n
+## Blocking Conditions
+
+- The primary flow fails.
+- A relevant build, check, or test fails without explanation.
+- Secrets or sensitive data would be exposed.
+- A high risk is not accepted by the user.
+- Verification evidence is missing for the risk level.
+

package/ai-orchestration-ecosystem/protocols/performance-budgets.md

@@ -0,0 +1,41 @@
+# Protocol: Performance Budgets
+
+## Objective
+
+Define measurable performance limits for web, mobile, backend, and AI features.
+
+## Web Budgets
+
+- Initial load time.
+- Largest Contentful Paint.
+- Interaction latency.
+- JavaScript bundle size.
+- Image/media size.
+- API response time.
+
+## Mobile Budgets
+
+- Cold start time.
+- Screen transition latency.
+- Memory usage.
+- Battery/network impact.
+- Offline behavior.
+
+## Backend Budgets
+
+- p95/p99 latency.
+- Throughput expectations.
+- Queue/job processing time.
+- Database query limits.
+
+## Output
+
+```md
+Performance budget:
+Target flow:
+Metric:
+Threshold:
+Measurement method:
+Risk if exceeded:
+```
+

package/ai-orchestration-ecosystem/protocols/post-delivery-monitoring.md

@@ -0,0 +1,32 @@
+# Protocol: Post Delivery Monitoring
+
+## Objective
+
+Define how Sagaz should handle Post Delivery Monitoring while keeping the process clear, low-token, safe, and verifiable.
+
+## Required Practice
+
+- Start from the user goal and current project state.
+- Load only relevant context.
+- Separate facts, assumptions, inferences, risks, and decisions.
+- Ask permission before meaningful state changes.
+- Record evidence and residual risk.
+
+## Standard Recommendation Format
+
+```md
+Recommendation:
+Why now:
+What changes:
+Benefit:
+Risk:
+Permission required:
+```n
+## Blocking Conditions
+
+- The primary flow fails.
+- A relevant build, check, or test fails without explanation.
+- Secrets or sensitive data would be exposed.
+- A high risk is not accepted by the user.
+- Verification evidence is missing for the risk level.
+

package/ai-orchestration-ecosystem/protocols/production-readiness.md

@@ -0,0 +1,32 @@
+# Protocol: Production Readiness
+
+## Objective
+
+Define how Sagaz should handle Production Readiness while keeping the process clear, low-token, safe, and verifiable.
+
+## Required Practice
+
+- Start from the user goal and current project state.
+- Load only relevant context.
+- Separate facts, assumptions, inferences, risks, and decisions.
+- Ask permission before meaningful state changes.
+- Record evidence and residual risk.
+
+## Standard Recommendation Format
+
+```md
+Recommendation:
+Why now:
+What changes:
+Benefit:
+Risk:
+Permission required:
+```n
+## Blocking Conditions
+
+- The primary flow fails.
+- A relevant build, check, or test fails without explanation.
+- Secrets or sensitive data would be exposed.
+- A high risk is not accepted by the user.
+- Verification evidence is missing for the risk level.
+

package/ai-orchestration-ecosystem/protocols/quality-gates.md

@@ -0,0 +1,32 @@
+# Protocol: Quality Gates
+
+## Objective
+
+Define how Sagaz should handle Quality Gates while keeping the process clear, low-token, safe, and verifiable.
+
+## Required Practice
+
+- Start from the user goal and current project state.
+- Load only relevant context.
+- Separate facts, assumptions, inferences, risks, and decisions.
+- Ask permission before meaningful state changes.
+- Record evidence and residual risk.
+
+## Standard Recommendation Format
+
+```md
+Recommendation:
+Why now:
+What changes:
+Benefit:
+Risk:
+Permission required:
+```n
+## Blocking Conditions
+
+- The primary flow fails.
+- A relevant build, check, or test fails without explanation.
+- Secrets or sensitive data would be exposed.
+- A high risk is not accepted by the user.
+- Verification evidence is missing for the risk level.
+

package/ai-orchestration-ecosystem/protocols/release-strategy.md

@@ -0,0 +1,38 @@
+# Protocol: Release Strategy
+
+## Objective
+
+Release changes safely using rollout strategies appropriate to risk.
+
+## Strategies
+
+- Preview environment.
+- Feature flag.
+- Canary release.
+- Phased rollout.
+- Blue/green deployment.
+- Immediate rollback.
+- Hotfix release.
+
+## Required Checks
+
+- Define release scope.
+- Define rollout method.
+- Define rollback criteria.
+- Define monitoring checks after release.
+- Confirm migrations and environment variables.
+- Confirm communication needs when users are affected.
+
+## Output
+
+```md
+Release plan:
+Scope:
+Strategy:
+Pre-release checks:
+Rollout steps:
+Rollback criteria:
+Post-release monitoring:
+Permission required:
+```
+

package/ai-orchestration-ecosystem/protocols/secure-sdlc.md

@@ -0,0 +1,38 @@
+# Protocol: Secure SDLC
+
+## Objective
+
+Build security into every phase of development instead of treating it as a final review.
+
+## Required Checks
+
+- Security requirements.
+- Threat modeling for meaningful risk.
+- Secure design review.
+- Secure implementation practices.
+- Secrets scanning.
+- Dependency vulnerability scanning.
+- Static analysis when available.
+- Authentication and authorization review.
+- Input validation and output encoding review.
+- Security verification before production.
+
+## Threat Modeling Prompt
+
+```md
+Assets:
+Actors:
+Trust boundaries:
+Entry points:
+Threats:
+Mitigations:
+Residual risk:
+```
+
+## Blocking Conditions
+
+- Secrets committed or exposed.
+- Authentication or authorization risk is unresolved.
+- Sensitive data is handled without a clear protection model.
+- High-risk threat has no mitigation or explicit user acceptance.
+

package/ai-orchestration-ecosystem/protocols/squad-pipeline-handoffs.md

@@ -0,0 +1,32 @@
+# Protocol: Squad Pipeline Handoffs
+
+## Objective
+
+Define how Sagaz should handle Squad Pipeline Handoffs while keeping the process clear, low-token, safe, and verifiable.
+
+## Required Practice
+
+- Start from the user goal and current project state.
+- Load only relevant context.
+- Separate facts, assumptions, inferences, risks, and decisions.
+- Ask permission before meaningful state changes.
+- Record evidence and residual risk.
+
+## Standard Recommendation Format
+
+```md
+Recommendation:
+Why now:
+What changes:
+Benefit:
+Risk:
+Permission required:
+```n
+## Blocking Conditions
+
+- The primary flow fails.
+- A relevant build, check, or test fails without explanation.
+- Secrets or sensitive data would be exposed.
+- A high risk is not accepted by the user.
+- Verification evidence is missing for the risk level.
+

package/ai-orchestration-ecosystem/protocols/sre-readiness.md

@@ -0,0 +1,41 @@
+# Protocol: SRE Readiness
+
+## Objective
+
+Ensure production systems are designed for reliability, operations, incident response, and recovery.
+
+## When To Use
+
+Use for production web apps, mobile backends, APIs, critical automations, paid products, systems with real users, or anything with meaningful availability expectations.
+
+## Required Checks
+
+- Define expected availability, latency, and reliability targets.
+- Identify critical user journeys.
+- Define service level indicators and service level objectives when appropriate.
+- Define an error budget or acceptable failure threshold when appropriate.
+- Define incident severity levels.
+- Define escalation and owner responsibilities.
+- Create or update operational runbooks.
+- Ensure logs, metrics, traces, or equivalent diagnostics exist.
+- Define rollback or mitigation paths.
+- Define post-incident review expectations.
+
+## Recommendation Format
+
+```md
+Reliability recommendation:
+Why now:
+Critical user journey:
+Reliability target:
+Operational risk:
+Runbook needed:
+Permission required:
+```
+
+## Blocking Conditions
+
+- No way to detect failure in a production-critical flow.
+- No rollback or mitigation for high-risk release.
+- No owner or response path for production incidents.
+

package/ai-orchestration-ecosystem/protocols/stack-selection.md

@@ -0,0 +1,32 @@
+# Protocol: Stack Selection
+
+## Objective
+
+Define how Sagaz should handle Stack Selection while keeping the process clear, low-token, safe, and verifiable.
+
+## Required Practice
+
+- Start from the user goal and current project state.
+- Load only relevant context.
+- Separate facts, assumptions, inferences, risks, and decisions.
+- Ask permission before meaningful state changes.
+- Record evidence and residual risk.
+
+## Standard Recommendation Format
+
+```md
+Recommendation:
+Why now:
+What changes:
+Benefit:
+Risk:
+Permission required:
+```n
+## Blocking Conditions
+
+- The primary flow fails.
+- A relevant build, check, or test fails without explanation.
+- Secrets or sensitive data would be exposed.
+- A high risk is not accepted by the user.
+- Verification evidence is missing for the risk level.
+

package/ai-orchestration-ecosystem/protocols/testing-matrix.md

@@ -0,0 +1,32 @@
+# Protocol: Testing Matrix
+
+## Objective
+
+Define how Sagaz should handle Testing Matrix while keeping the process clear, low-token, safe, and verifiable.
+
+## Required Practice
+
+- Start from the user goal and current project state.
+- Load only relevant context.
+- Separate facts, assumptions, inferences, risks, and decisions.
+- Ask permission before meaningful state changes.
+- Record evidence and residual risk.
+
+## Standard Recommendation Format
+
+```md
+Recommendation:
+Why now:
+What changes:
+Benefit:
+Risk:
+Permission required:
+```n
+## Blocking Conditions
+
+- The primary flow fails.
+- A relevant build, check, or test fails without explanation.
+- Secrets or sensitive data would be exposed.
+- A high risk is not accepted by the user.
+- Verification evidence is missing for the risk level.
+

package/ai-orchestration-ecosystem/quickstart.md

@@ -0,0 +1,25 @@
+# Quickstart
+
+Use this file as the minimum entry point for operating Sagaz with low token usage.
+
+## Modes
+
+- `solo`: one primary agent executes with minimal gates.
+- `squad`: the orchestrator coordinates specialized agents.
+- `factory`: create a new product with specs, implementation, and verification.
+- `audit`: review an existing project for bugs, risks, regressions, and missing tests.
+- `refactor`: change internals while preserving behavior.
+
+## Standard Flow
+
+1. Clarify objective, constraints, definition of done, and risks.
+2. Select the smallest sufficient workflow and squad.
+3. Create specifications when scope or risk is meaningful.
+4. Execute in small verifiable steps.
+5. Test proportionally to risk.
+6. Record decisions, evidence, residual risks, and next steps.
+
+## Permission Policy
+
+Ask one question at a time only when the answer changes architecture, scope, cost, risk, or definition of done. When uncertainty is small, make a conservative assumption, record it, and continue.
+

package/ai-orchestration-ecosystem/skills/exhaustive-verification.md

@@ -0,0 +1,14 @@
+# Skill: Exhaustive Verification
+
+## When To Use
+
+Use this when the task benefits from Exhaustive Verification.
+
+## Procedure
+
+1. Define expected behavior.
+2. Identify risks and invariants.
+3. Execute in small steps.
+4. Verify results.
+5. Document evidence and residual risks.
+

package/ai-orchestration-ecosystem/skills/refactor-proofing.md

@@ -0,0 +1,14 @@
+# Skill: Refactor Proofing
+
+## When To Use
+
+Use this when the task benefits from Refactor Proofing.
+
+## Procedure
+
+1. Define expected behavior.
+2. Identify risks and invariants.
+3. Execute in small steps.
+4. Verify results.
+5. Document evidence and residual risks.
+

package/ai-orchestration-ecosystem/skills/spec-first-development.md

@@ -0,0 +1,14 @@
+# Skill: Spec First Development
+
+## When To Use
+
+Use this when the task benefits from Spec First Development.
+
+## Procedure
+
+1. Define expected behavior.
+2. Identify risks and invariants.
+3. Execute in small steps.
+4. Verify results.
+5. Document evidence and residual risks.
+

package/ai-orchestration-ecosystem/squads/code-audit.md

@@ -0,0 +1,45 @@
+# Squad: Code Audit
+
+## Use
+
+Use this squad when the task matches Code Audit.
+
+## Agents
+
+- Orchestrator
+- Product Strategist when product decisions are needed
+- Technology Strategist when stack or deployment decisions are needed
+- Software Architect when architecture matters
+- Design Studio roles when UI exists
+- Implementation Engineer when code changes are needed
+- QA Verifier for verification
+- Security Governance when data, auth, secrets, payments, or production are involved
+- GitHub Ops when versioning, PRs, releases, or checks are useful
+
+## Flow
+
+1. Confirm objective, constraints, definition of done, and risks.
+2. Select relevant tasks and protocols.
+3. Execute the phase in small verifiable steps.
+4. Produce evidence and residual risks.
+5. Ask permission before handing off to the next team.
+
+## Required Gates
+
+- Clear objective and definition of done.
+- Relevant specification or task contract.
+- Handoff approval between major phases.
+- Verification evidence proportional to risk.
+- Residual risks documented.
+
+## Standard Output
+
+```md
+Scope:
+Current phase:
+Evidence:
+Risks:
+Next team:
+Permission request:
+```
+

package/ai-orchestration-ecosystem/squads/design-studio.md

@@ -0,0 +1,45 @@
+# Squad: Design Studio
+
+## Use
+
+Use this squad when the task matches Design Studio.
+
+## Agents
+
+- Orchestrator
+- Product Strategist when product decisions are needed
+- Technology Strategist when stack or deployment decisions are needed
+- Software Architect when architecture matters
+- Design Studio roles when UI exists
+- Implementation Engineer when code changes are needed
+- QA Verifier for verification
+- Security Governance when data, auth, secrets, payments, or production are involved
+- GitHub Ops when versioning, PRs, releases, or checks are useful
+
+## Flow
+
+1. Confirm objective, constraints, definition of done, and risks.
+2. Select relevant tasks and protocols.
+3. Execute the phase in small verifiable steps.
+4. Produce evidence and residual risks.
+5. Ask permission before handing off to the next team.
+
+## Required Gates
+
+- Clear objective and definition of done.
+- Relevant specification or task contract.
+- Handoff approval between major phases.
+- Verification evidence proportional to risk.
+- Residual risks documented.
+
+## Standard Output
+
+```md
+Scope:
+Current phase:
+Evidence:
+Risks:
+Next team:
+Permission request:
+```
+