ai-squad 0.1.0

package/sections/opencode/workflow/sessions.md

@@ -0,0 +1,20 @@
+# Session Management
+
+## Session Logs
+Each development session is recorded in `.agent/sessions/` as `YYYY-MM-DD.md`.
+
+## When to Write
+At the end of each session, the `squad-memory-keeper` writes a session log.
+
+## Format
+```markdown
+# Session YYYY-MM-DD
+- **Agent:** [which agent ran]
+- **Tasks worked:** [task IDs with status changes]
+- **Decisions made:** [links to decision log entries]
+- **Files changed:** [key files modified]
+- **Next steps:** [what to do in the next session]
+```
+
+## Purpose
+Session logs provide continuity across sessions so agents can pick up where they left off.

package/sections/opencode/workflow/state.md

@@ -0,0 +1,20 @@
+# ${PROJECT_NAME} — Current State
+
+## Active Tasks
+[List active tasks with status]
+
+## Recently Completed
+[List recently completed tasks]
+
+## Blocked Items
+[List blocked items with reason]
+
+## Quality Gates
+| Gate | Status |
+|------|--------|
+| Tests | — |
+| Lint | — |
+| Build | — |
+
+## Next Session Priority
+1. [Most important next task]

package/sections/opencode/workflow/tasks.md

@@ -0,0 +1,13 @@
+# Task Management
+
+## Finding the Current Task
+Check `state.md` for active tasks. Open the task file at `.agent/tasks/`.
+
+## Updating Task Status
+When starting work, mark status `in_progress`. When done, mark `completed`. Blocked tasks should note the blocker reason.
+
+## Creating New Tasks
+The PO agent (`squad-po`) creates tasks in `.agent/tasks/` using `task-template.md` as the base. Each task gets a numeric ID (e.g., `task-01.md`). The PO validates acceptance criteria are concrete and verifiable.
+
+## Task Lifecycle
+`pending` → `in_progress` → (reviewed by `squad-reviewer`) → `completed`

package/sections/shared/frontmatter/adversarial-reviewer.yaml

@@ -0,0 +1,7 @@
+name: squad-adversarial-reviewer
+description: >-
+  Adversarial Reviewer for ${PROJECT_NAME}. Second-pass review hunting for missed bugs.
+  Use after the first reviewer has passed the code.
+mode: subagent
+hidden: true
+color: "#EF4444"

package/sections/shared/frontmatter/architect.yaml

@@ -0,0 +1,7 @@
+name: squad-architect
+description: >-
+  System Architect for ${PROJECT_NAME}. Designs systems, data flows, type contracts.
+  Use for system design, architecture decisions.
+mode: subagent
+hidden: true
+color: "#8B5CF6"

package/sections/shared/frontmatter/docs-writer.yaml

@@ -0,0 +1,7 @@
+name: squad-docs-writer
+description: >-
+  Documentation Writer for ${PROJECT_NAME}. Maintains project context documentation.
+  Use for updating project docs and context files.
+mode: subagent
+hidden: true
+color: "#06B6D4"

package/sections/shared/frontmatter/fullstack-dev.yaml

@@ -0,0 +1,7 @@
+name: squad-fullstack-dev
+description: >-
+  Full-Stack Developer for ${PROJECT_NAME}. Implements features end-to-end.
+  Use for implementing features, fixing bugs, or writing code.
+mode: subagent
+hidden: true
+color: "#4A90D9"

package/sections/shared/frontmatter/memory-keeper.yaml

@@ -0,0 +1,7 @@
+name: squad-memory-keeper
+description: >-
+  Memory Keeper for ${PROJECT_NAME}. Manages project memory across sessions.
+  Use for recording decisions, conventions, and session logs.
+mode: subagent
+hidden: true
+color: "#10B981"

package/sections/shared/frontmatter/po.yaml

@@ -0,0 +1,7 @@
+name: squad-po
+description: >-
+  Product Owner for ${PROJECT_NAME}. Creates and manages task files.
+  Use for scoping features, creating tasks, or validating requirements.
+mode: subagent
+hidden: true
+color: "#D4845F"

package/sections/shared/frontmatter/qa.yaml

@@ -0,0 +1,7 @@
+name: squad-qa
+description: >-
+  QA Engineer for ${PROJECT_NAME}. Tests implementations and verifies acceptance criteria.
+  Use after implementation, before closing tasks.
+mode: subagent
+hidden: true
+color: "#F59E0B"

package/sections/shared/frontmatter/reviewer.yaml

@@ -0,0 +1,7 @@
+name: squad-reviewer
+description: >-
+  Code Reviewer for ${PROJECT_NAME}. GREEN/RED review of completed code.
+  Use after implementation is complete.
+mode: subagent
+hidden: true
+color: "#50C878"

package/sections/shared/frontmatter/teamlead.yaml

@@ -0,0 +1,8 @@
+name: squad-teamlead
+description: >-
+  ${PROJECT_NAME} Team Lead. Orchestrates the squad. Only agent that talks to the user.
+  Assesses work type, dispatches agents in the right pipeline, synthesizes results.
+  Delegation-only — never writes code, never fixes issues directly.
+mode: primary
+hidden: false
+color: "#D4845F"

package/sections/shared/frontmatter/ux-ui.yaml

@@ -0,0 +1,7 @@
+name: squad-ux-ui
+description: >-
+  UX/UI Reviewer for ${PROJECT_NAME}. Reviews UI against design standards.
+  Use for reviewing UI implementation quality.
+mode: subagent
+hidden: true
+color: "#EC4899"

package/sections/shared/memory/conventions.md

@@ -0,0 +1,10 @@
+# Convention Log — ${PROJECT_NAME}
+
+Discovered and established conventions that agents should follow. Appended over time as patterns emerge.
+
+## Format
+
+### YYYY-MM-DD — Convention Name
+Description of the convention and why it exists.
+
+---

package/sections/shared/memory/decisions.md

@@ -0,0 +1,13 @@
+# Decision Log — ${PROJECT_NAME}
+
+Architecture Decision Records. Each entry records a significant technical decision, why it was made, and its consequences.
+
+## Format
+
+### YYYY-MM-DD — Decision Title
+- **Context:** Why this decision was needed
+- **Decision:** What was decided
+- **Alternatives considered:** What other options were evaluated
+- **Consequences:** Positive and negative outcomes of this decision
+
+---

package/sections/shared/memory/patterns.md

@@ -0,0 +1,12 @@
+# Pattern Library — ${PROJECT_NAME}
+
+Reusable code patterns and architectural templates discovered in the codebase.
+
+## Format
+
+### Pattern Name
+- **Use when:** Description of when to apply this pattern
+- **Structure:** Code or diagram showing the pattern
+- **Example:** Reference to where this pattern is used in the codebase
+
+---

package/sections/shared/task-template.md

@@ -0,0 +1,44 @@
+---
+id: task-XX
+status: pending
+priority: medium
+assigned: squad-fullstack-dev
+---
+
+# Task XX: [Task Title]
+
+## Product Context
+- Problem / user need: [What problem, for whom, and why now]
+- Desired outcome: [The user or business outcome this task should create]
+- Success measure: [Observable metric, behavior, or validation signal]
+- Decision owner: [Person or role accountable for material product decisions]
+
+## Priority
+[Critical / High / Medium / Low] — [Why this should be done now, including relevant trade-offs]
+
+## Scope
+What this task covers. Be specific about the boundaries.
+
+## Out of Scope
+Explicit exclusions to prevent scope creep.
+
+## Acceptance Criteria
+1. [Concrete, verifiable statement]
+2. [Given/When/Then format preferred]
+
+## Technical Notes
+- Relevant files: path/to/file.ts
+- Patterns to follow: [reference to existing convention]
+- Constraints: [API limits, performance targets, etc.]
+
+## Dependencies
+- Task IDs that must be completed before this one
+
+## Assumptions & Risks
+- Assumptions: [Confirmed assumptions that materially affect the task]
+- Risks / open decisions: [Risk, unresolved decision, owner, and required action]
+
+## Delivery & Follow-up
+- Rollout / migration / rollback: [Required approach, or "Not applicable"]
+- Observability / support / documentation: [Required work, or "Not applicable"]
+- Outcome review: [When and how to assess the success measure, or "Not applicable"]