Menu
OMA skill index (SLASH-FIRST) — /oh-my-agy:autopilot five-phase + ralph, ultrawork, team, verify, setup; CLI is optional ledger only
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
In-session OMA autonomous delivery — invoke /oh-my-agy:autopilot with a goal; run deep-interview→ralplan→ultragoal→code-review→ultraqa HERE (no terminal required)
In-session OMA cancel — invoke /oh-my-agy:cancel; stop active modes HERE, leave resume-friendly state (CLI stop optional)
In-session OMA merge-readiness gate — invoke /oh-my-agy:code-review; write APPROVE+CLEAR or REQUEST CHANGES under .agy/reviews HERE
In-session OMA requirements gate — invoke /oh-my-agy:deep-interview; Socratic clarify HERE, write .agy/specs (no terminal first)
In-session OMA persistence loop — invoke /oh-my-agy:ralph; keep going until PRD stories pass with verify evidence (CLI optional)
In-session OMA planning gate — invoke /oh-my-agy:ralplan; author+steelman+critic HERE, write .agy/plans (no terminal first)
| name | oma-runtime |
| description | OMA skill index (SLASH-FIRST) — /oh-my-agy:autopilot five-phase + ralph, ultrawork, team, verify, setup; CLI is optional ledger only |
You are running under oh-my-agy (OMA) — the Antigravity / Claude Code / Grok orchestration layer (sibling of OMC / OMX).
User habit matches Claude Code /autopilot: work happens inside the agent session.
| Rule | Detail |
|---|---|
| Canonical form | /oh-my-agy:<skill-name> (namespaced) |
Bare /autopilot | Left to OMC (or host) if present — do not steal it |
| Happy path | No SID / CID / revision required to start |
CLI oma … | Optional durable outer ledger only — never “open a terminal first” |
When a skill is invoked via slash or skill protocol, treat $ARGUMENTS as the goal/seed and execute HERE.
| User intent | Canonical slash | Skill body | Optional CLI helper |
|---|---|---|---|
| Full autonomous delivery (OMX five-phase) | /oh-my-agy:autopilot | skills/autopilot/SKILL.md | oma autopilot … (ledger only) |
| Clarify requirements | /oh-my-agy:deep-interview | skills/deep-interview/SKILL.md | (phase of autopilot) |
| Consensus-style plan gate | /oh-my-agy:ralplan | skills/ralplan/SKILL.md | oma autopilot consensus (optional) |
| Durable implement + verify | /oh-my-agy:ultragoal | skills/ultragoal/SKILL.md | handoff/advance (optional); oma team explicit |
| Merge-readiness review | /oh-my-agy:code-review | skills/code-review/SKILL.md | oma autopilot review (optional) |
| Adversarial QA gate | /oh-my-agy:ultraqa | skills/ultraqa/SKILL.md | oma autopilot qa (optional) |
| Persist until verified done | /oh-my-agy:ralph | skills/ralph/SKILL.md | oma ralph -- <task> (optional) |
| Parallel independent work | /oh-my-agy:ultrawork | skills/ultrawork/SKILL.md | oma ultrawork -- <task> (optional) |
| Read-only research | /oh-my-agy:search | skills/search/SKILL.md | oma search -- <query> (optional) |
| Multi-worker tmux team | /oh-my-agy:team | skills/team/SKILL.md | oma team start|status|tick|… |
| Stop active modes | /oh-my-agy:cancel | skills/cancel/SKILL.md | cancel managed session if bound |
| Evidence before “done” | /oh-my-agy:verify | skills/verify/SKILL.md | tests/build/doctor evidence |
| Install/enable plugin | /oh-my-agy:setup | skills/setup/SKILL.md | oma setup / oma doctor |
| This index | /oh-my-agy:oma-runtime | (this file) | oma skill list / oma skill show <name> |
deep-interview → ralplan → ultragoal → code-review → ultraqa → complete
Invoke end-to-end with:
/oh-my-agy:autopilot <goal>
The autopilot skill runs each phase in this conversation, writing artifacts under .agy/. It does not require oma autopilot drive to begin.
When a managed launch includes <<<OMA_SKILL_PROTOCOL …>>>, that protocol is mandatory for the run. Autopilot drive (optional) injects the skill for the current phase when the user maintains an outer ledger.
| Phase | Path |
|---|---|
| Specs | .agy/specs/ |
| Plans | .agy/plans/ |
| Implement ledger | .agy/ultragoal/ |
| Reviews | .agy/reviews/ |
| QA | .agy/qa/ |
| Ralph PRD | .agy/ralph/ |
| Autopilot mirror | .agy/autopilot/state.json (optional) |
Do not invent .omc / .omx paths unless migrating.
AGENTS.md unless the user explicitly requests a merge policy change.OMA_SESSION_ID, OMA_LAUNCH_NONCE, OMA_INVOCATION_GENERATION.git reset --hard / git clean -fd.spawn/spawnSync argv arrays in any tooling you write — no shell exec.skills/verify).ralplan with reason — no silent scope reduction.OMA_STATE_ROOT or platform default) when CLI is used..agy/ artifacts written by the phase skills.oma CLI surfaceFor operators who want a durable outer ledger (not the happy path):
oma skill list
oma skill show autopilot
oma autopilot start -- "<goal>"
oma autopilot status --session <id>
oma autopilot drive --session <id> --conversation <cid> --expected-revision <n>
oma autopilot handoff --session <id> --expected-revision <n> --key <key> --path <artifact>
oma autopilot advance --session <id> --expected-revision <n> --evidence <file>
oma autopilot consensus --session <id> --expected-revision <n> --role architect|critic --verdict approve|revise --note <text>
oma autopilot review --session <id> --expected-revision <n> --evidence <file>
oma autopilot qa --session <id> --expected-revision <n> --evidence <file>
oma autopilot return-ralplan --session <id> --expected-revision <n> --reason <text>
oma ralph -- "<task>"
oma setup && oma doctor --no-strict-plugin
Keep CLI handoffs consistent with existing .agy/ artifacts when both are used. In-session slash remains primary.