namba-ai: Skill for Claude Code

State effect: mutating workflow entry point. Use help/probe paths read-only, and otherwise expect repository state or GitHub state to change.

Generated instruction contract for this command skill:

• Purpose: keep the role or command scope explicit, bounded, and testable.
• Boundary: honor read-only versus mutating state effects, configured sandbox mode, and assigned file or workflow ownership.
• Required output: report concrete actions, changed paths or artifacts, validation evidence, and pass/fail status or blockers.
• Pass/fail criteria: claim success only when acceptance criteria and configured validation are satisfied; otherwise name the exact blocker and impact.
• Evidence expectations: cite source artifacts such as SPEC files, .namba/ configs, diffs, test output, PR/check links, or generated manifests instead of relying on unsupported assertions.
• Security responsibilities: never expose or commit secrets; treat auth, privacy, destructive commands, permission changes, and external network or credential use as security-sensitive.
• Destructive command and escalation policy: do not run destructive commands unless explicitly requested; request approval for privileged, networked, or sandbox-blocked actions only when the active approval mode allows it, and otherwise report the blocker or use a safe non-escalating path.
• Fallback implementer boundary: if a specialist path is unavailable and the main/default implementer takes over, stay within the assigned scope and preserve the same evidence and validation duties.
• Portability: keep durable guidance non-project-specific unless the current repository config or SPEC explicitly provides the project detail.

Use this skill when the user explicitly says $namba-plan-review, asks to create a SPEC and run the full pre-implementation review loop, or wants namba plan plus the review flow bundled into one skill.

Behavior:

• Resolve the target SPEC from an explicit SPEC-XXX; otherwise create the next SPEC with namba plan for feature work, namba harness for reusable agent/skill/workflow/orchestration work, or namba fix --command plan for bugfix planning.
• When creating a new SPEC from a user request, inherit the same clarification gate as $namba-plan: ask first and do not run the CLI while the raw request is still vague, short, or missing Goal/Scope/Constraints/Acceptance.
• Prefer the installed namba CLI for the SPEC-creation step when it is available; keep .namba/ as the source of truth if you need to do the setup manually.
• Inherit the same safe-by-default planning branch contract as namba plan: create or switch to the dedicated spec/... branch in the current workspace by default, treat --current-workspace as the explicit current-branch escape hatch, and do not create planning worktrees.
• Read .namba/specs/<SPEC>/spec.md, plan.md, and acceptance.md before launching reviews or revising the planning artifacts.
• Launch product, engineering, and design review passes in parallel when subagent routing is available, using $namba-plan-pm-review, $namba-plan-eng-review, and $namba-plan-design-review as the three authoritative review tracks.
• Prefer namba-product-manager, namba-planner, and namba-designer for the three review tracks, and use namba-plan-reviewer as the aggregate validator when custom-agent routing is available.
• After the three review tracks finish, run an aggregate validation pass over spec.md, plan.md, acceptance.md, and .namba/specs/<SPEC>/reviews/*.md to check coverage gaps, contradictions, and whether the advisory readiness state is credible.
• If the aggregate validator finds issues, revise the SPEC or review artifacts directly, rerun only the affected review tracks, and repeat the validation loop instead of restarting every pass blindly.
• Refresh .namba/specs/<SPEC>/reviews/readiness.md after each review and validation cycle so the advisory summary stays current.
• Before ending, inspect .namba/specs/<SPEC>/reviews/readiness.md; if it does not say Cleared reviews: 3/3, do not report the review loop as complete. Continue the review loop when useful, or make the final next-work item explicitly name $namba-plan-review SPEC-XXX or the missing individual review skills to run next.
• Keep the loop bounded and explicit: stop when the readiness state is clear enough to proceed or when the remaining blockers are concrete enough that another loop would be redundant.
• Keep the whole flow advisory by default; missing depth or blockers should be visible, not silently converted into a hard gate.