Skip to content

j-mauro-r/SAT-Dengue-V2 / .agents/skills/deepworkplan/addons/devcontainer

deepworkplan-addon-devcontainer

Verified

Optional DeepWorkPlan addon that gives a repository a reproducible dev container through devcontainer-kit (the `dck` command, DailybotHQ/devcontainer-kit pinned at v1.0.1, interface 2) - a vendor-neutral thin integrator that detects the kit with `dck doctor --json`, offers the kit and its tag-pinned `dck-dockerfile` skill, and lets them render the repository's own layout (.devcontainer/devcontainer.json, docker/local/<service>/Dockerfile, docker/local/docker-compose.yaml, dev.sh) from the official runtime image pinned by digest - no shared base image required. Reconciles an existing layout and never clobbers it, keeps loopback ports and agent forwarding, and validates the result with a real build. Opt-in, never required, never a conformance gate.

FileKindSize
SKILL.mdskill md22.9 KB
SPEC.mdmarkdown15.3 KB
addon.jsonconfig326 B
templates/INTEGRATION.mdmarkdown4.0 KB

templates/INTEGRATION.md

View on GitHub
# Template — devcontainer through `dck` (reason, don't copy-paste)

Reason every option from the repository's **real** files; never from a
preset you remember. Detected reality wins. When the `dck-dockerfile` skill
is installed, it does this detection and asks once for confirmation; this
page is the same reasoning for when it is not.

## 1. Stack → runtime

| Signal in the repo | Runtime |
|---|---|
| `package.json` / `pnpm-lock.yaml` (Node/TS app, service or lambda) | `node`, at the version `engines` declares |
| `pyproject.toml` / `uv.lock`, `requirements.txt`, `setup.py`, `Pipfile` | `python` |
| neither (Go, Rust, shell, docs, mixed) | `debian` (add the toolchain in the Dockerfile's project section, outside the kit's managed blocks) |

The Dockerfile starts from that runtime's **official image**, pinned by
digest in the kit's pin file. No shared base image is needed.

## 2. Service, ports, backing services

- Service name: the existing compose service, else the repository
  directory name.
- Ports: only ports the app really listens on in dev (existing compose,
  framework config, dev-script flags); they bind `127.0.0.1`.
- Backing services (database, cache, queue) the app really uses in dev go in
  `docker/local/docker-compose.yaml` **outside** the kit's managed blocks,
  with their versions pinned.

## 3. What goes in the image

| Piece | When |
|---|---|
| coding-agent CLIs through `ak install` | only the kinds the person names; each CLI home is a named volume so logins survive `dev.sh rebuild`. Agents launch in autonomy by default — the container is the sandbox; the compose's commented `AGENTKIT_PERMISSIONS=ask` line is the opt-out |
| DeepWorkPlan Vim | default on; off when the person never edits inside |
| extras (for example chromium for Lighthouse or Playwright) | only when the repository's own scripts need them |
| the Dailybot CLI | **only** when the `dailybot` addon is enabled and asks for it |

## 4. Herdr container profile (optional)

Want the container as a Herdr machine? Turn on the kit's Herdr machine
option; `bash dev.sh up` registers it. Running `dck herdr add` edits
`~/.ssh/config` (a guarded include) — ask for that separately. Git and SSH
inside use the host's agent, never a mounted `~/.ssh` unless the developer
turns on `[ssh] copy_keys` (below); the kit picks the
agent socket by Docker provider. Your own `~/.ssh/config` aliases for git
hosting services work inside too (`ssh_host_config`, default on: public keys
and trusted host keys only, no private key copied); `dck up` may offer
`ssh-add` for a missing key — the developer answers. Hosts beyond the git
services are the developer's own opt-in through `[ssh] host_extra` in their
host profile (never the repository's `dck.toml`); never set it for them.
The same holds for `[ssh] copy_keys` (kit `v0.3.2`): the developer's host
profile only, for tools that read key files; while on, the whole host
`~/.ssh` is readable inside and private keys rest in the state volume.
Several services in one compose project? Give each its own
`[compose] state_volume` (`dck init --state-volume NAME`).
Inside, peers use the pinned herdr-peers skill (`../../herdr/install.md`). `[herdr] mesh` (default `true`, Docker
Desktop only) connects the repository's container to the other dck
containers; it widens trust between them, so name that cost, and set
`mesh = false` when the container should stay on its own.

## 5. Render and validate

```bash
dck init --dry-run <options from §1–§3>     # plan + diffs; writes nothing
dck init <options from §1–§3>               # after acceptance (per-file prompt on a terminal)
dck doctor --json                           # interface 2, runtime, repo, drift
bash dev.sh build && bash dev.sh up         # a real build; a failure is a failed run
dck exec <service> -- <the repo's real test command>
```

Use the option names the kit's `v1.0.1` documentation gives (`dck init
--help`). Record each outcome. A failure is a finding about the environment,
never a repository conformance failure.
writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deploymentStars: 322kUpdated Claude CodeCodexVerifiedInstall
xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release.Stars: 118kUpdated Claude CodeCodexVerifiedInstall