원클릭으로 Manus에서 모든 스킬 실행

setup-autonomous-fleet

스타0

포크0

업데이트2026년 7월 5일 07:42

User-invoked setup only — do not auto-activate. Configure a repo for autonomous-fleet: runtime adapter, branch prefix, default campaign bundle, optional community installs. Run when the user says setup autonomous fleet, configure fleet, or first fleet run on a repo.

설치

Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.

Manus에서 실행

출처

ravidsrk

ravidsrk/autonomous-fleet

GitHub 저장소 열기 Creator 저장소 보기

다운로드

Manus에서 실행

관련 직업SOC

SOC 직업 분류 기준

소프트웨어 개발자컴퓨터 및 수학직·SOC 15-1252

파일 탐색기

5 개 파일

SKILL.md

readonly

이 저장소의 다른 Skills

같은 저장소

autonomous-fleet-core

ravidsrk/autonomous-fleet

The portable, tool-agnostic ENGINE for running fully-autonomous multi-agent engineering jobs. Shipped mission skills (doc-sync, test-coverage, adversarial-review-and-fix) invoke THIS engine plus exactly one ADAPTER (orca, claude-code, grok, or another runtime). Eighteen additional missions are documented under docs/exploratory/missions/ and re-promote on real-run evidence. This core holds everything that does NOT depend on orchestration tool: self-orientation, fully-autonomous coordinator behaviour with file-ledger boolean gates, context-handoff to survive compaction, the worker-placement DECISION LOGIC (dependent vs independent), the PR-per-task pipeline with commits-preserved + conflict-aware merge + worktree cleanup, the empirical risk tiers, safety rails, secret hygiene, and commit/authorship policy. It speaks in PRIMITIVES; the ACTIVE ADAPTER maps each primitive to its tool's real commands. Load with a mission skill and one runtime adapter — do not run alone.

2026-07-080

devex-audit

ravidsrk/autonomous-fleet

[Tier 2 · moderate autonomy · full review gate · developer-experience] Run a live developer experience audit with frozen DX scorecard and ranked documentation gaps. Maps gstack /devex-review and /plan-devex-review into fleet ledgers with optional document-generate follow-ups. Use when onboarding friction, CLI help, or docs TTHW need evidence before ship. Trigger on: "DX audit", "developer experience review", "devex mission", "audit onboarding", "score the getting started flow", "plan devex review".

2026-07-050

landing-page-convergence

ravidsrk/autonomous-fleet

[Tier 2 · moderate autonomy · full review gate] Force a production landing page that has DIVERGED from an approved design into full fidelity with that design - section by section, with a named divergence checklist as the forcing function so it converges instead of drifting. Use when a live page no longer matches its design export and needs to be brought back into line (a single page/site, not a whole app - that's design-integration). Extracts the design system, diffs production against it, and closes each divergence as its own PR. Runs via the autonomous-fleet-core engine. Trigger on: "make the landing page match the design", "the production page diverged from the design", "bring the landing page into parity", "fix the landing page to match the mockup".

2026-07-050

legacy-rebuild

ravidsrk/autonomous-fleet

[Tier 3 · high blast radius · expect rework · run deliberately, review the floor + architecture] Adversarially review a legacy app, research the best current architecture, and rebuild it end to end on a modern foundation while PRESERVING everything it currently does. Use when an app uses outdated everything (old deps, poor architecture, e.g. JS-inlined-in-HTML, no build/module system) and needs a real modernization, not a patch. Incremental and shippable per PR — NOT a big-bang rewrite — rebuilt against a captured behaviour floor so nothing is silently lost. Runs via the autonomous-fleet-core engine. Trigger on: "rebuild this legacy app", "modernize the whole codebase", "this uses all old versions, rebuild it properly", "re-architect end to end".

2026-07-050

product-framing

ravidsrk/autonomous-fleet

[Tier 2 · moderate autonomy · full review gate · pre-build] Turn vague product intent into a frozen, review-approved product spec before any implementation. Runs gstack-style office-hours forcing questions plus CEO/design/eng plan reviews (via optional community skills), then emits docs/product-spec.md as the single source of truth for downstream build missions. Use when the user has an idea, wedge, or feature direction but no frozen spec yet. Trigger on: "frame this product", "office hours on this idea", "is this worth building", "freeze the product spec", "run plan reviews before we code", "product framing mission".

2026-07-050

release-document

ravidsrk/autonomous-fleet

[Tier 2 · moderate autonomy · full review gate · post-ship docs] Post-ship documentation sweep and deploy verification checklist after a version lands. Maps gstack /document-release, /ship, /land-and-deploy, and /canary into fleet frozen artifacts. Use after merge or deploy when changelog, user docs, and canary notes need closure. Trigger on: "document this release", "post-ship docs", "release documentation mission", "update docs after ship", "canary doc sweep".

2026-07-050

name	setup-autonomous-fleet
description	User-invoked setup only — do not auto-activate. Configure a repo for autonomous-fleet: runtime adapter, branch prefix, default campaign bundle, optional community installs. Run when the user says setup autonomous fleet, configure fleet, or first fleet run on a repo.
license	MIT
compatibility	Requires git; gh CLI recommended for PR workflows
metadata	{"author":"ravidsrk","version":"1.1.3","fleet-component":"setup"}

Setup autonomous-fleet

Scaffold per-repo fleet configuration so coordinators, fleet-program, and missions start with consistent defaults. Prompt-driven — explore, confirm with the user, then write.

Process

1. Explore

Read what already exists; do not assume:

git rev-parse --show-toplevel, git remote -v, default branch
CLAUDE.md / AGENTS.md — existing agent sections?
docs/agents/fleet-config.md — prior setup output?
DECISIONS.md — recorded BRANCH_PREFIX or adapter choice?
.agents/skills/ or host skill dirs — which fleet skills are installed?
gh auth status — PR workflow available?

2. Present findings — one section at a time

Walk the user through the sections sequentially — A–C are decisions, D is an optional install, E is detect-and-record only. Each section: short explainer, choices, default.

Section A — Runtime adapter.

Which host will run fleet coordinators? This picks the adapter skill that maps engine primitives to real commands.

Choice	Adapter skill
Orca (recommended)	`autonomous-fleet-adapter-orca`
Grok Build	`autonomous-fleet-adapter-grok`
Claude Code	`autonomous-fleet-adapter-claude-code`
OpenAI Codex	`autonomous-fleet-adapter-codex`

Default: Orca — reference runtime with structural build-blind review and native orchestration. Use another adapter only when Orca is unavailable.

Section B — Branch prefix.

Fleet missions branch as <prefix><task-slug>. Default derives from git user.name (slugified) or fleet/.

Confirm override or accept default (fleet/ or slugified maintainer name).

Section C — Default bundle.

When intent is vague, which fleet-program preset should the umbrella suggest?

Bundle	Preset	When
Fix my repo	`repo-health`	Docs/tests/cleanup pass
Ship safely	`ship-with-proof`	Pre-merge hardening
Finish product	`align-then-ship`	Tier 3 — explicit only
Production check	`quality-gate`	Readiness without full doc-sync
Single mission	`none`	User always names a mission

Default: repo-health.

Section D — Community skills (optional).

Third-party skills attach as Optional/Worker only — see community-skills.md. Install per bundle, not the full catalog. Gstack-derived missions declare community-recommends with mode: warn; preflight-community.sh prints hints when bundles are absent.

Ask which bundle ids apply (gstack-browser, gstack-security, gstack, agent-skills, mattpocock). Map the default campaign preset (Section C) to bundles when obvious (e.g. quality-gate → gstack-browser). Record choices in fleet-config.md; do not install without user consent.

When the user confirms, print or run:

./scripts/install-community.sh <bundle-id> --dry-run
# after explicit consent:
./scripts/install-community.sh <bundle-id> --execute --host <adapter-host>

Section D2 — Authorship mode.

How should fleet commits record agent involvement? Default attributed: commits are authored as the maintainer AND carry a Co-Authored-By trailer naming the agent — transparent provenance. Choose maintainer-only only if this repo's contribution policy forbids trailers; that choice is recorded.

Section E — Substrate path (no user decision; detect and record).

Resolve where the Python verification substrate lives and record it as SUBSTRATE_PATH in fleet-config.md:

Framework clone (./scripts/validate_run_archive.py exists) → scripts/.
Skills-install → .agents/skills/autonomous-fleet-core/assets/substrate/ (ships with autonomous-fleet-core; version pinned in its substrate-manifest.json).
Neither present → record SUBSTRATE_PATH: none and note that the four verification layers run as prose-only disciplines until the core skill is installed.

Offer to install the substrate's Python deps into the repo's environment (python3 -m pip install -r <SUBSTRATE_PATH>/requirements.txt) — ask before touching any environment.

Also probe for a published docs-site toolchain (docusaurus.config., mkdocs.yml, docs/conf.py, starlight in astro.config.) and record LEDGER_DIR accordingly (.fleet/docs/ when found, else docs/) — fleet ledgers must never break or publish through a site build.

Also probe which research tools exist on this host (monid on PATH, a Context7/library-docs MCP, a deep-research-class skill) and record them as RESEARCH_TOOLS in fleet-config.md; record native-web when none are found (engine-workers.md RESEARCH DISCIPLINE resolves through this field first).

3. Confirm draft

Show draft of:

docs/agents/fleet-config.md (see references/fleet-config-template.md)
## Autonomous fleet block for CLAUDE.md or AGENTS.md

Let the user edit before writing.

4. Write

Pick host doc: edit CLAUDE.md if present, else AGENTS.md, else ask which to create.

Update ## Autonomous fleet in-place if it exists — no duplicate blocks.

Write docs/agents/fleet-config.md.

Append a one-line setup record to DECISIONS.md with adapter, prefix, bundle, date.

5. Install skills (if missing)

First detect install mode — the helper scripts only exist in a framework clone, not in a skills-installed repo (npx skills copies only skills/ into .agents/skills/, never scripts/):

if [ -f ./scripts/install-skills.sh ]; then echo clone; else echo skills-install; fi

Skills-install mode (no ./scripts/) — the common case. Print npx skills commands; do not reference ./scripts/*, which is not shipped here:

npx skills@1.5.12 add https://github.com/ravidsrk/autonomous-fleet \
  --skill setup-autonomous-fleet \
  --skill autonomous-fleet \
  --skill autonomous-fleet-core \
  --skill fleet-program \
  --skill autonomous-fleet-adapter-<chosen> \
  --skill doc-sync \
  -y
# or every skill: --skill '*'

Clone mode (./scripts/install-skills.sh present) — contributor path:

./scripts/install-skills.sh --all
# or minimal:
./scripts/install-skills.sh autonomous-fleet fleet-program autonomous-fleet-core \
  autonomous-fleet-adapter-<chosen> doc-sync

For community bundles, print install commands from skills/autonomous-fleet-core/references/community-skills.md.

6. Verify

Skills-install mode (no ./scripts/) — there is no ./scripts/run-campaign.sh to call. Confirm the install instead: the listed skills now resolve under .agents/skills/, and the chosen adapter is among them. Drive the first run interactively from the agent's chat (/setup-autonomous-fleet is done; invoke a mission or fleet-program next).

Clone mode — suggest a headless dry-run:

./scripts/run-campaign.sh <adapter-runtime> --preset <bundle> --dry-run

After setup

Coordinators read docs/agents/fleet-config.md during SELF-ORIENTATION (override defaults there over engine defaults when present). Re-run this skill when changing adapter, prefix, or default bundle.