deploy-status
Show deployed git SHAs across configured production targets and compare them against origin/main. Use during deploy verification, incident response, and release drift checks. Triggers on: "what is deployed", "deploy status", "production drift", "is prod on main", "compare prod to main".
- License 许可
- MIT
- 最近更新
- 2026-10-26
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- deploy-status
- description
- Show deployed git SHAs across configured production targets and compare them against origin/main. Use during deploy verification, incident response, and release drift checks. Triggers on: "what is deployed", "deploy status", "production drift", "is prod on main", "compare prod to main".
- allowed-tools
- Read Grep Glob Bash
- argument-hint
- [--dry-run] [--no-fetch]
- user-invocable
- true
- disable-model-invocation
- false
02 资源文件 5
- 参考文档examples/deploy-targets.example.json924 B
- 参考文档references/aws-amplify.md566 B
- 参考文档references/fly.md1.1 KB
- 参考文档references/vercel.md1.0 KB
- 脚本scripts/deploy-ops.mjs44.5 KB
03 SKILL.md 正文
Deploy Status
Report what git revision is live across production deploy targets and whether
each target matches origin/main.
Workflow
-
Resolve the helper script:
if [ -f "$HOME/.claude/skills/deploy-status/scripts/deploy-ops.mjs" ]; then DEPLOY_OPS="$HOME/.claude/skills/deploy-status/scripts/deploy-ops.mjs" elif [ -f "skills/deploy-status/scripts/deploy-ops.mjs" ]; then DEPLOY_OPS="skills/deploy-status/scripts/deploy-ops.mjs" else echo "deploy-status helper not found; re-run dotbabel bootstrap or dotbabel init" >&2 exit 2 fi -
Run the status command from the consuming project root:
node "$DEPLOY_OPS" status $ARGUMENTS -
Preserve the helper's exit code:
0- every target is in sync withorigin/main1- at least one target is behind, ahead, or otherwise drifted2- target discovery, provider auth, provider parsing, or SHA detection failed
Target Discovery
Discovery is configuration-first only where auto-discovery cannot prove a target. The helper:
- Auto-discovers Vercel from
.vercel/project.json. - Auto-discovers Fly.io from top-level
app = "..."infly.toml. - Merges
.claude/deploy-targets.jsonwhen present. Config entries override matching auto-discovered targets and add non-discoverable platforms.
Example config: examples/deploy-targets.example.json.
Smoke checks
A target in .claude/deploy-targets.json may declare a smoke array, run
after a deploy to confirm the thing actually answers:
node "$DEPLOY_OPS" smoke # human-readable
node "$DEPLOY_OPS" smoke --json # machine-readable, see schemas/dotbabel.smoke-report.schema.json
node "$DEPLOY_OPS" smoke --dry-run # list the checks without running them
Exit codes match the status command: 0 every check passed (or none is
declared, or it was a dry run — neither is a failure), 1 at least one failed
or the run hit its time budget, 2 target discovery failed.
Read verdict rather than inferring success from the exit code: pass,
fail, not_configured (no target declares checks) and not_run (a dry run).
A dry run also sets dry_run: true and reports each planned check with
ok: null, so a stray --dry-run in a pipeline cannot be mistaken for a green
gate over zero executed checks.
Two check types. An http check is GET only and retries up to 3 times with
2s, 4s and 8s backoff; a command check runs its argv exactly once,
because a command may not be idempotent and a retry would repeat a side effect
nobody agreed to. Each request is aborted after 10 seconds — a timeout is
transient, so it is retried on the backoff schedule — and the whole run stops
after 300 seconds.
A check with no expect_status defaults to requiring 2xx. A response
arriving is not the same as a response being good: without that default, a
check declaring no expectations reported green against a hard 500, which turns
an outage into a passing release gate. Set expect_status explicitly when you
want a specific non-2xx code.
These requests go to production and may carry a real credential, so the guards are enforced rather than advised:
- A URL with embedded credentials is refused, and is not echoed back in the error — the config carrying it is already a committed secret.
httpsis required for every host exceptlocalhostand127.0.0.1.- Redirects are followed manually, at most 3, and only to the same https origin. Handing the response the power to choose the next host is how a secret header reaches somewhere it was never meant to go.
- A header value is read only from a named environment variable
(
{ "env": "SMOKE_BEARER" }). A literal in the config is refused, and a variable that is unset fails the check rather than quietly sending no header and passing against an anonymous response. - Header names are reported; values never are. The failing response body is shown truncated and redacted, because an endpoint that echoes the header back would otherwise leak it through the very message reporting the failure.
- A refusal from any of these guards is never retried: a retry cannot turn it into a pass, and a differing later response would mask it.
{
"kind": "vercel",
"project": "my-app",
"smoke": [
{
"type": "http",
"url": "https://my-app.example.com/healthz",
"expect_status": 200,
"expect_text": "ok"
},
{
"type": "http",
"url": "https://my-app.example.com/api/me",
"expect_status": 200,
"headers": { "Authorization": { "env": "SMOKE_BEARER" } }
},
{ "type": "command", "argv": ["./scripts/smoke-api.sh", "--prod"] }
]
}
Provider References
Load only the provider notes that match discovered targets:
| Provider | Reference |
|---|---|
| Vercel | references/vercel.md |
| Fly.io | references/fly.md |
| AWS Amplify | references/aws-amplify.md |
Rules
- Do not prompt for provider tokens. Use the existing CLI auth state
(
vercel whoami,fly auth whoami). - Do not deploy or roll back from this skill. This skill is read-only except for
git fetch origin main --quiet. - If a provider cannot expose a git SHA, report that target as unknown and exit
2rather than inventing drift data. - On a smoke failure, recommend
/rollback-prodand stop. The rule floor forbids production changes without explicit instruction, so this skill never rolls back on its own.
04 同仓库更多 Skills
yolo
Start temporary autonomy for the current task or an explicit scope with a required --for expiry. Codex permits scoped pushes, merges, deploys, and releases after strict testing and validation; Full access is a separate runtime setting. Claude Code uses a session permission hook with a protected-action denylist.
veracity-audit
Audit a data pipeline for Veracity and Value. Dispatches data-scientist, compliance-auditor, and data-engineer agents with project context injected at dispatch time. All source paths are supplied via flags — no defaults, no project assumptions. Subcommands: audit (full pipeline walk), score-check (scoring math only), gate-check (gate coverage only), source-trace <label> (single source end-to-end). Invoke when: "audit pipeline", "veracity check", "scoring math correct?", "check gates", "trace source", "verify quality gates", "formula correct?".
validate-spec
Audit an already-implemented spec against the codebase. Walks each constraint (ARCH-N, PERF-N, KD-N, etc.) and acceptance criterion, grounds findings in file:line evidence, runs the spec.json acceptance_commands, and writes a single audit doc to docs/audits/. Use whenever the user asks to "validate a spec", "audit a spec", "check if spec is implemented", "verify the spec is done", "is this spec really done", or otherwise wants closure on spec-driven work. Read-only against the spec — produces an audit, never modifies the spec itself.
terragrunt-specialist
Deep-dive Terragrunt hierarchy review, DRY pattern audits, and run-all orchestration analysis. Use for structured investigations of multi-environment Terragrunt layouts, dependency graphs, remote state config, and hook correctness. Triggers on: "Terragrunt audit", "run-all review", "dependency block", "DRY pattern review", "env hierarchy audit", "mock_outputs", "terragrunt hooks".
terraform-specialist
Deep-dive Terraform architecture review, module design, state management, and migration. Use for structured investigations of Terraform workspaces, provider configuration, module coupling, import workflows, and test coverage. Triggers on: "Terraform audit", "module review", "state management", "Terraform import", "workspace design", "provider config review", "Terraform testing".
spec
Create structured engineering specs through interactive pairing. Use when the user wants to create a spec, design doc, technical specification, architecture document, RFC, or engineering plan. Triggers on "let's spec this out", "create a spec", "design doc", "write a technical plan", "plan the architecture". Works for greenfield and brownfield projects. Outputs to docs/specs/.