← Browse

@stevegjones/deploy-team

A

Build a team Docker image from a manifest in .archon/teams/. Resolves plugin paths, generates Dockerfile with additive copy-only, makes plugins read-only, bakes team CLAUDE.md.

skillclaude

Install

agr install @stevegjones/deploy-team --target claude

Writes 1 file into .claude/skills/, pinned to git-cfe3a78e.

  • .claude/skills/deploy-team/SKILL.md

Document


name: deploy-team description: Build a team Docker image from a manifest in .archon/teams/. Resolves plugin paths, generates Dockerfile with additive copy-only, makes plugins read-only, bakes team CLAUDE.md. disable-model-invocation: false argument-hint: "--name [--image <registry/path>]"

Deploy Team Image

Build a team-specific Docker image from a team manifest.

Arguments

  • --name <team-name> (required) — team manifest in .archon/teams/<team-name>.yaml
  • --image <tag> (optional) — custom image tag (default: sdlc-worker:<team-name>)

Steps

1. Read and validate the manifest

Read .archon/teams/<team-name>.yaml. Validate:

python3 ${CLAUDE_PLUGIN_ROOT}/scripts/validate_team_manifest.py \
    .archon/teams/<team-name>.yaml \
    --installed-plugins "${CLAUDE_PLUGINS_DIR:-$HOME/.claude/plugins}/installed_plugins.json" \
    --project-root .

If validation fails, report errors and stop. Check status field — refuse to build inactive or decommissioned teams.

2. Verify base and full images exist

docker image inspect sdlc-worker:base >/dev/null 2>&1
docker image inspect sdlc-worker:full >/dev/null 2>&1

If missing, inform user and offer to build them:

bash ${CLAUDE_PLUGIN_ROOT}/docker/build-base.sh
bash ${CLAUDE_PLUGIN_ROOT}/docker/build-full.sh

3. Generate team CLAUDE.md

Use the generate_team_claude_md.py script to create the team's CLAUDE.md from the manifest:

python3 ${CLAUDE_PLUGIN_ROOT}/scripts/generate_team_claude_md.py \
    .archon/teams/<team-name>.yaml \
    --installed-plugins "${CLAUDE_PLUGINS_DIR:-$HOME/.claude/plugins}/installed_plugins.json" \
    --output .archon/teams/.generated/<team-name>-CLAUDE.md

If the project has a CLAUDE.md, concatenate it (project first, team second).

4. Build the team image

bash ${CLAUDE_PLUGIN_ROOT}/docker/build-team.sh <team-name> [--image <tag>]

This generates a Dockerfile with additive copy-only (no prune) and builds the image. The generated Dockerfile and CLAUDE.md go to .archon/teams/.generated/.

5. Update manifest timestamp

After successful build, update image_built in the manifest YAML to current ISO-8601 datetime.

6. Report

Report the following:

  • Image name and tag
  • Image size (from docker image inspect)
  • Agent count and skill count (from manifest)
  • Generated file locations (Dockerfile and CLAUDE.md)

Trustgrade A

  • passBody integrity

    Whether the stored document is plausibly the kind of file the artifact declares, rather than something fetched by mistake.

  • passType matchnot applicable to this artifact type

    Whether the artifact is really the kind of thing its metadata claims it is.

  • passFreshness

    How long since the source repository was last pushed to.

  • passPrompt injection

    Scans the artifact's own text for instructions aimed at your agent rather than at you.

  • passLicense

    Whether the source repository declares an SPDX license permissive enough to redistribute.

How the grade is calculated

Each check contributes 0 points when it passes, 1 when it warns, and 2 when it fails. The total maps to a letter:

  • Aevery check passed
  • Bone warning
  • Ctwo warnings
  • Dprompt injection or body integrity failed, or three warnings
  • Fone of those failed, and something else is wrong

These are automated hygiene checks, not a security audit, and not a dependency or vulnerability scan. A grade of A means nothing was flagged — not that the artifact is safe.

Versions

  • git-cfe3a78e3ec32026-07-31