Preamble

Environment Detection

mkdir -p ~/.omni-skills/sessions
_PROACTIVE=$(~/.claude/skills/superomni/bin/config get proactive 2>/dev/null || echo "true")
_BRANCH=$(git branch --show-current 2>/dev/null || echo "unknown")
_TEL_START=$(date +%s)
echo "Branch: $_BRANCH | PROACTIVE: $_PROACTIVE"

PROACTIVE Mode

If PROACTIVE is false: do NOT proactively suggest skills. Only run skills the user explicitly invokes. If you would have auto-invoked, say: "I think [skill-name] might help here — want me to run it?" and wait.

Completion Status Protocol

Report status using one of these at the end of every skill session:

• DONE — All steps completed. Evidence provided.
• DONE_WITH_CONCERNS — Completed with issues. List each concern explicitly.
• BLOCKED — Cannot proceed. State what blocks you and what was tried.
• NEEDS_CONTEXT — Missing information. State exactly what you need.

Auto-Advance Rule

Pipeline stage order: THINK → PLAN → REVIEW → BUILD → VERIFY → SHIP → REFLECT

REVIEW is the only human gate. All other stages auto-advance on DONE.

Status	At REVIEW stage	At all other stages
DONE	STOP — present review summary, wait for user input (Y / N / revision notes)	Auto-advance — print [STAGE] DONE → advancing to [NEXT-STAGE] and immediately invoke next skill
DONE_WITH_CONCERNS	STOP — present concerns, wait for user decision	STOP — present concerns, wait for user decision
BLOCKED / NEEDS_CONTEXT	STOP — present blocker, wait for user	STOP — present blocker, wait for user

When auto-advancing:

1. Write the session artifact to docs/superomni/
2. Print: [STAGE] DONE → advancing to [NEXT-STAGE] ([skill-name])
3. Immediately invoke the next pipeline skill

Session Continuity

When the user sends a follow-up message after a completed session, before doing anything else:

1. Scan for prior session context:

   ls docs/superomni/specs/spec-*.md docs/superomni/plans/plan-*.md docs/superomni/ .superomni/ 2>/dev/null | head -20
   git log --oneline -3 2>/dev/null
   

   To find the latest spec or plan:

   _LATEST_SPEC=$(ls docs/superomni/specs/spec-*.md 2>/dev/null | sort | tail -1)
   _LATEST_PLAN=$(ls docs/superomni/plans/plan-*.md 2>/dev/null | sort | tail -1)
   

2. If context exists → re-engage the skill framework. Pick the skill that matches the current stage (see workflow skill for stage → skill mapping) and announce: "Continuing in superomni mode — picking up at [stage] using [skill-name]."
3. If no context → treat as a fresh session and offer the relevant skill from the Quick Reference table in using-skills/SKILL.md.

Question Confirmation Protocol

When asking the user a question, match the confirmation requirement to the complexity of the response:

Question type	Confirmation rule
Single-choice — user picks one option (A/B/C, 1/2/3, Yes/No)	The user's selection IS the confirmation. Do NOT ask "Are you sure?" or require a second submission.
Free-text input — user types a value and presses Enter	The submitted text IS the confirmation. No secondary prompt needed.
Multi-choice — user selects multiple items from a list	After the user lists their selections, ask once: "Confirm these selections? (Y to proceed)" before acting.
Complex / open-ended discussion — back-and-forth clarification	Collect all input, then present a summary and ask: "Ready to proceed with the above? (Y/N)" before acting.

Rule: never add a redundant confirmation layer on top of a single-choice or text-input answer.

Custom Input Option Rule: Whenever you present a predefined list of choices (A/B/C, numbered options, etc.), always append a final "Other" option that lets the user describe their own idea:

  [last letter/number + 1]) Other — describe your own idea: ___________

When the user selects "Other" and provides their custom text, treat that text as the chosen option and proceed exactly as you would for any other selection. If the custom text is ambiguous, ask one clarifying question before proceeding.

Context Window Management

Load context progressively — only what is needed for the current phase:

Phase	Load these	Defer these
Planning	Latest docs/superomni/specs/spec-*.md, constraints, prior decisions	Full codebase, test files
Implementation	Latest docs/superomni/plans/plan-*.md, relevant source files	Unrelated modules, docs
Review/Debug	diff, failing test output, minimal repro	Full history, specs

If context pressure is high: summarize prior phases into 3-5 bullet points, then discard raw content.

Output Directory

All skill artifacts are written to docs/superomni/ (relative to project root). See the Document Output Convention in CLAUDE.md for the full directory map.

Feedback Signal Protocol

Agent failures are harness signals — not reasons to retry the same approach:

• 1 failure → retry with a different approach
• 2 failures → surface to user: "Tried [A] and [B], both failed. Recommend [C]."
• 3 consecutive failures → STOP. Report BLOCKED. Treat as a harness deficiency signal. Recommended: invoke harness-engineering skill to update the harness before retrying.
• Uncertain about security → STOP and report NEEDS_CONTEXT
• Scope exceeds verification capacity → STOP and flag blast radius

It is always OK to stop and say "this is too hard for me." Escalation is expected, not penalized.

Performance Checkpoint

After completing any skill session, run a 3-question self-check before writing the final status:

1. Process — Did I follow all defined phases? If any were skipped, state why.
2. Evidence — Is every claim backed by a test result, command output, or file reference? If not, gather the missing evidence now.
3. Scope — Did I stay within the task boundary? If I touched files outside the original scope, flag them explicitly.

If any answer is NO, address it before reporting DONE. If it cannot be addressed, report DONE_WITH_CONCERNS and name the gap.

For a full performance evaluation spanning the entire sprint, use the self-improvement skill.

Telemetry (Local Only)

_TEL_END=$(date +%s)
_TEL_DUR=$(( _TEL_END - _TEL_START ))
~/.claude/skills/superomni/bin/analytics-log "SKILL_NAME" "$_TEL_DUR" "OUTCOME" 2>/dev/null || true

Nothing is sent to external servers. Data is stored only in ~/.omni-skills/analytics/.

Vibe — Framework Entry Point

Goal: Activate the superomni skill framework, detect the current pipeline stage, and guide the user to the right skill.

Usage

/vibe              — activate framework, detect stage, show guided menu
/vibe status       — show current pipeline position and available next steps
/vibe reset        — clear superomni artifacts and restart from THINK

Iron Law: One Entry Point, Full Pipeline

/vibe is the single unified entry point. It never executes work itself — it detects the current stage and delegates to the appropriate skill.

CRITICAL: Do NOT use EnterPlanMode

When /vibe is invoked, you MUST follow this skill's phases (Phase 1-4) directly. NEVER trigger Claude Code's built-in EnterPlanMode tool at any point during the vibe workflow or any subsequent superomni skill session. EnterPlanMode bypasses the superomni pipeline entirely. For planning, delegate to the writing-plans skill. For implementation, delegate to executing-plans or subagent-development.

Phase 1: Detect Current Pipeline Stage

Scan for existing artifacts to determine where the project is in the sprint pipeline:

# Artifact detection (glob for dynamic filenames)
_HAS_SPEC=$(ls docs/superomni/specs/spec-*.md 2>/dev/null | sort | tail -1)
_HAS_PLAN=$(ls docs/superomni/plans/plan-*.md 2>/dev/null | sort | tail -1)
_HAS_EXECUTIONS=$(ls docs/superomni/executions/*.md 2>/dev/null | head -1)
_HAS_IMPROVEMENTS=$(ls docs/superomni/improvements/*.md 2>/dev/null | head -1)

# Session matching: extract session from plan filename for review detection
if [ -n "$_HAS_PLAN" ]; then
  _PLAN_SESSION=$(basename "$_HAS_PLAN" .md | sed 's/plan-[^-]*-//' | sed 's/-[0-9]*$//')
  _HAS_MATCHING_REVIEW=$(ls docs/superomni/reviews/review-*-${_PLAN_SESSION}-*.md 2>/dev/null | head -1)
  _PLAN_OPEN=$(grep -c '^\- \[ \]' "$_HAS_PLAN" 2>/dev/null || echo "0")
  _PLAN_DONE=$(grep -c '^\- \[x\]' "$_HAS_PLAN" 2>/dev/null || echo "0")
fi

# Recent git activity
git log --oneline -5 2>/dev/null
git status --short 2>/dev/null

Stage Detection Matrix

Use the following priority-ordered rules (first match wins):

Priority	Condition	Stage	Skill
1	No artifacts at all	THINK	brainstorm
2	spec-*.md exists, no plan-*.md	PLAN	writing-plans
3	plan-*.md exists, no review doc matching its session	REVIEW	plan-review — human gate: wait for Y/N
4	Plan reviewed + approved, has open items (- [ ])	BUILD	executing-plans
5	plan-*.md all checked	VERIFY	Required: code-review → qa → verification. Optional: security-audit (if security-relevant), production-readiness (if deploying)
6	Verified	SHIP	ship, finishing-branch
7	Shipped	REFLECT	self-improvement → retro (run sequentially in one stage)

Session matching for REVIEW detection: Extract the [session] segment from the plan filename (e.g., plan-main-auth-refactor-20260404.md → session = auth-refactor). A matching review doc must contain the same session identifier (e.g., review-main-auth-refactor-*.md). If no matching review exists → stage is REVIEW.

Auto-Advance Rule

REVIEW is the only human gate. All other stages auto-advance on DONE.

• Non-REVIEW stages with DONE: Auto-invoke the next skill immediately. Print: [STAGE] DONE → advancing to [NEXT-STAGE] ([skill-name])
• REVIEW stage with DONE: STOP. Present the review summary and ask: Plan review complete. Proceed to BUILD? (Y / N / describe revisions)
• Any stage with DONE_WITH_CONCERNS / BLOCKED / NEEDS_CONTEXT: STOP and wait for user.

Only show the guided menu when:

• No clear next step can be determined
• The user invoked /vibe without arguments

If the user passes arguments with /vibe (e.g., /vibe I want to build a CLI tool), treat the arguments as the starting prompt and route to the detected skill with that context.

Phase 2: Show Guided Menu

Present the available commands only when auto-advance does not apply:

| Command | What it does |
|---------|-------------|
| /brainstorm | Design a feature — produces spec-[branch]-[session]-[date].md |
| /write-plan | Turn a spec into an executable plan |
| /execute-plan | Run the plan step by step |
| /review | Plan review (before BUILD) |
| /qa | Quality assurance and test coverage |
| /verify | Verify task completion |
| /production-readiness | Pre-deploy readiness check |
| /investigate | Exploratory investigation |
| /self-improve | Post-task performance evaluation |
| /retro | Engineering retrospective |
| /workflow | See the full sprint pipeline |
| /ship | Release workflow |

Suggested next step → [skill-name]: [reason based on detected stage]

Phase 3: Handle Subcommands

/vibe status

Run the stage detection from Phase 1 and display:

Pipeline: THINK → PLAN → REVIEW → BUILD → VERIFY → SHIP → REFLECT
Stage: [current] | Branch: [branch]
Artifacts: spec-*.md [Y/N] | plan-*.md [Y/N] | executions [N] | reviews [N] | prod-readiness [N] | improvements [N]
Next → [skill-name]: [reason]

/vibe reset

Warn before clearing artifacts:

WARNING: This will remove all superomni artifacts:
  - docs/superomni/specs/spec-*.md
  - docs/superomni/plans/plan-*.md
  - docs/superomni/executions/
  - docs/superomni/reviews/
  - docs/superomni/subagents/
  - docs/superomni/production-readiness/

Self-improvement history will be preserved:
  - docs/superomni/improvements/
  - docs/superomni/evaluations/
  - docs/superomni/harness-audits/

Proceed? (Y/N)

If confirmed:

rm -f docs/superomni/specs/spec-*.md docs/superomni/plans/plan-*.md
rm -rf docs/superomni/executions/ docs/superomni/reviews/ docs/superomni/subagents/ docs/superomni/production-readiness/
echo "Reset complete. Starting fresh from THINK stage."

Then re-run Phase 1-2 (will detect THINK stage).

Phase 4: Delegate to Skill

After displaying the banner and menu:

1. If the user provided arguments → invoke the detected skill with those arguments
2. If no arguments → wait for user input
3. When the user chooses a command or describes what they want to do → invoke the matching skill

Important: /vibe never executes implementation work directly. It always delegates to the appropriate skill. Do NOT use Claude Code's built-in EnterPlanMode — always delegate to the superomni skill (e.g., writing-plans for planning).

Report status: DONE — framework activated, stage detected, user guided to next skill.