Claude Code Team Rollout
Portability: 4 stdlib scripts, no network, no LLM calls. They read config and counts; they change nothing.
One engineer's Claude Code setup is a preference. A team's is a policy. Settings
merge from five scopes, permissions.allow arrays combine across files, and
deny beats ask beats allow wherever it is written, so nobody can read the
effective policy from a single file. This skill makes it readable, then checks it.
Workflow
# 1 - Guardrails: lint shared, local and managed settings together.
python scripts/settings_linter.py --repo . --managed /etc/claude-code/managed-settings.json
# 2 - Inventory: every MCP server, plugin and marketplace, against your approved list.
python scripts/mcp_plugin_inventory.py --repo . --allowlist approved.json
# 3 - Instructions: what CLAUDE.md actually loads, layer by layer.
python scripts/claude_md_layering_audit.py --repo .
# 4 - Adoption: each member runs this locally and shares only the JSON counts.
python scripts/adoption_tracker.py --member alice --output json > alice.json
python scripts/adoption_tracker.py --merge *.json --seats 12 --weeks-since-launch 6
Start from assets/managed-settings.example.json:
bypass mode disabled, secrets denied, risky commands on ask, managed hooks
only, an MCP and marketplace allowlist. Ship it through MDM before the first
seat; policy added after habits form becomes a negotiation.
What blocks (exit 4)
- Blanket shell allows (
Bash, Bash(*)) and pre-approved destructive commands
(force push, rm -rf, sudo, pipe-to-shell, publish, infra apply).
- A literal credential in a shared settings
env block, a committed .mcp.json, or a committed CLAUDE.md.
.claude/settings.local.json committed to git.
- An MCP server, plugin or marketplace missing from the approved list.
Warnings cover dead rules (an allow that a deny shadows, mcp__* in allow),
enableAllProjectMcpServers, unpinned npx servers, plain-http remotes,
network-calling hooks, broken or too-deep @imports, duplicated or
contradictory instructions, and resident CLAUDE.md size.
Hard rules
- Policy lives in managed settings. Project files cannot enable bypass mode
or telemetry, and a team guardrail in a file anyone can edit is a suggestion.
- Approve MCP servers by name, URL or command, never by
enableAllProjectMcpServers.
- Pin what runs. Unpinned
npx/uvx servers and unpinned marketplaces change under you.
- Never collect transcripts. Share the adoption JSON (counts only); transcripts are plaintext and can hold secrets.
- Report findings, not verdicts on people. Adoption is a team metric; inactive seats are a rollout question, not a performance one.
- Settings keys change fast. An unknown key is INFO; check the settings reference before calling it a typo.
Scripts
| Script |
Role |
Exit codes |
scripts/settings_linter.py |
S1-S13 across shared, local, user and managed scopes |
0 · 2 warn · 3 bad input · 4 blocker |
scripts/mcp_plugin_inventory.py |
One table of servers, plugins, marketplaces; M1-M6 |
0 · 2 · 3 · 4 |
scripts/claude_md_layering_audit.py |
Resident vs on-demand layers, imports (4 hops), L1-L11 |
0 · 2 · 3 · 4 |
scripts/adoption_tracker.py |
Member counts; team stage PILOT / SPREADING / EMBEDDED / STALLED |
0 · 2 stalled · 3 |
All support --sample and --output json.
References and assets
references/settings_and_permissions.md — scopes, precedence, rule syntax, managed-only keys (7 sources)
references/mcp_and_plugin_governance.md — MCP scopes, allowlists, marketplaces, pinning (6 sources)
references/memory_and_adoption.md — CLAUDE.md layering, transcripts, OpenTelemetry, adoption stages (6 sources)
assets/managed-settings.example.json — baseline team policy
assets/approved-components.example.json — allowlist format for step 2
assets/rollout_plan.md — four-phase plan with exit criteria per phase
1---2name: coding-agent-rollout3description: Use when rolling Claude Code out to a team or organization, or auditing a rollout already under way — writing or reviewing shared and managed settings.json permissions, deciding which MCP servers, plugins and marketplaces to approve, cleaning up layered CLAUDE.md files, or measuring whether the team actually uses it. Lints settings across scopes, inventories every MCP server and plugin against an approved list, resolves CLAUDE.md layers and imports, and turns local session data into privacy-safe adoption counts.4argument-hint: "[repo path | managed-settings.json | 'measure adoption']"5license: MIT6metadata:7 source: https://skillmd.com/skills/slipcast-dev/coding-agent-rollout8 version: 1.0.09 build_pattern: "Four-phase rollout (guardrails, inventory, instructions, adoption) with 4 deterministic stdlib scripts and blocking exit codes; every settings key and path verified against code.claude.com/docs"10 distinct_from: "claude-coach (coaches one person's usage); skill-doctor (grades sessions to improve installed skills); security-guidance (a hook that warns on risky code patterns); prompt-governance (prompts in production apps, not the coding agent's configuration)"11disable-model-invocation: true # skill-optimizer12---1314# Claude Code Team Rollout1516> **Portability:** 4 stdlib scripts, no network, no LLM calls. They read config and counts; they change nothing.1718One engineer's Claude Code setup is a preference. A team's is a policy. Settings19merge from five scopes, `permissions.allow` arrays combine across files, and20deny beats ask beats allow wherever it is written, so nobody can read the21effective policy from a single file. This skill makes it readable, then checks it.2223## Workflow2425```bash26# 1 - Guardrails: lint shared, local and managed settings together.27python scripts/settings_linter.py --repo . --managed /etc/claude-code/managed-settings.json28# 2 - Inventory: every MCP server, plugin and marketplace, against your approved list.29python scripts/mcp_plugin_inventory.py --repo . --allowlist approved.json30# 3 - Instructions: what CLAUDE.md actually loads, layer by layer.31python scripts/claude_md_layering_audit.py --repo .32# 4 - Adoption: each member runs this locally and shares only the JSON counts.33python scripts/adoption_tracker.py --member alice --output json > alice.json34python scripts/adoption_tracker.py --merge *.json --seats 12 --weeks-since-launch 635```3637Start from [`assets/managed-settings.example.json`](assets/managed-settings.example.json):38bypass mode disabled, secrets denied, risky commands on `ask`, managed hooks39only, an MCP and marketplace allowlist. Ship it through MDM before the first40seat; policy added after habits form becomes a negotiation.4142## What blocks (exit 4)4344- Blanket shell allows (`Bash`, `Bash(*)`) and pre-approved destructive commands45 (force push, `rm -rf`, `sudo`, pipe-to-shell, publish, infra apply).46- A literal credential in a shared settings `env` block, a committed `.mcp.json`, or a committed CLAUDE.md.47- `.claude/settings.local.json` committed to git.48- An MCP server, plugin or marketplace missing from the approved list.4950Warnings cover dead rules (an allow that a deny shadows, `mcp__*` in allow),51`enableAllProjectMcpServers`, unpinned `npx` servers, plain-http remotes,52network-calling hooks, broken or too-deep `@imports`, duplicated or53contradictory instructions, and resident CLAUDE.md size.5455## Hard rules56571. **Policy lives in managed settings.** Project files cannot enable bypass mode58 or telemetry, and a team guardrail in a file anyone can edit is a suggestion.592. **Approve MCP servers by name, URL or command, never by `enableAllProjectMcpServers`.**603. **Pin what runs.** Unpinned `npx`/`uvx` servers and unpinned marketplaces change under you.614. **Never collect transcripts.** Share the adoption JSON (counts only); transcripts are plaintext and can hold secrets.625. **Report findings, not verdicts on people.** Adoption is a team metric; inactive seats are a rollout question, not a performance one.636. **Settings keys change fast.** An unknown key is INFO; check the settings reference before calling it a typo.6465## Scripts6667| Script | Role | Exit codes |68|---|---|---|69| `scripts/settings_linter.py` | S1-S13 across shared, local, user and managed scopes | 0 · 2 warn · 3 bad input · 4 blocker |70| `scripts/mcp_plugin_inventory.py` | One table of servers, plugins, marketplaces; M1-M6 | 0 · 2 · 3 · 4 |71| `scripts/claude_md_layering_audit.py` | Resident vs on-demand layers, imports (4 hops), L1-L11 | 0 · 2 · 3 · 4 |72| `scripts/adoption_tracker.py` | Member counts; team stage PILOT / SPREADING / EMBEDDED / STALLED | 0 · 2 stalled · 3 |7374All support `--sample` and `--output json`.7576## References and assets7778- [`references/settings_and_permissions.md`](references/settings_and_permissions.md) — scopes, precedence, rule syntax, managed-only keys (7 sources)79- [`references/mcp_and_plugin_governance.md`](references/mcp_and_plugin_governance.md) — MCP scopes, allowlists, marketplaces, pinning (6 sources)80- [`references/memory_and_adoption.md`](references/memory_and_adoption.md) — CLAUDE.md layering, transcripts, OpenTelemetry, adoption stages (6 sources)81- [`assets/managed-settings.example.json`](assets/managed-settings.example.json) — baseline team policy82- [`assets/approved-components.example.json`](assets/approved-components.example.json) — allowlist format for step 283- [`assets/rollout_plan.md`](assets/rollout_plan.md) — four-phase plan with exit criteria per phase