Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/everywan-dev/claude-code-engineering/deploy-to-production-safelynpx skills add everywan-dev/claude-code-engineering --skill deploy-to-production-safelygit clone --depth 1 https://github.com/everywan-dev/claude-code-engineeringWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/everywan-dev/claude-code-engineering/deploy-to-production-safely)<a href="https://agentmods.dev/skills/everywan-dev/claude-code-engineering/deploy-to-production-safely"><img src="https://agentmods.dev/badge/skills/everywan-dev/claude-code-engineering/deploy-to-production-safely.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00034 | $0.01524 |
| Opus 5 | $0.00017 | $0.00762 |
| Sonnet 5 | $0.00007 | $0.00305 |
| Haiku 4.5 | $0.00003 | $0.00152 |
Grade A, and why
deploy-to-production-safely scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 151 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Deploy to production safely
A production change is not a command. It is a sequence, and the sequence has an order. Skipping a step does not make the change faster — it moves the cost to the moment you can least afford it.
The order
Do these in this order. Not in the order that feels natural.
| # | Step | Done when |
|---|---|---|
| 1 | Photograph the starting state | You have the output of the commands that describe the system before, pasted somewhere you can read later |
| 2 | Take a copy — and verify it | You have listed the contents of the copy, not just seen the file exist |
| 3 | Write the rollback | The exact commands that undo this are written down, before anything is applied |
| 4 | Rehearse cold | The tool's dry-run mode has been run and its output pasted |
| 5 | Apply | — |
| 6 | Verify with a check that can fail | You ran something that would have gone red if the change had not worked |
Step 3 comes before step 5 for one reason: the moment you need the rollback is the moment you are least able to write it.
Step 1 — Photograph the starting state
Whatever the change touches, capture it first: the list of processes, the current configuration, the row counts, the routing table, the version in use.
You are not doing this for the change. You are doing it for the argument three days later about whether the change caused the thing that broke.
Step 2 — A copy you have not read is not a copy
🔴 A dump nobody has opened is not a backup. It is a file of unknown size and unknown content that makes everyone feel safer than they are.
Verifying a copy means reading back from the copy itself:
- A database export → list the objects inside it and count them.
- A configuration file → open the saved copy and confirm the section you are about to change is present in it.
- A whole directory → count the entries in the archive, not on disk.
If the copy cannot be listed, it is not a backup and the change does not proceed.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 151 lines · 34 tokens per session scan A 2ec7dc41d83c
deploy-to-production-safely is a skill published in the GitHub repository everywan-dev/claude-code-engineering (2 stars, last pushed 14d ago), licensed Apache-2.0. It adds 34 tokens to every session and 1,524 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
claude-project-manifest
Creates annotated bibliography-style manifests for Claude projects, tracking files, conversation threads, and relationships with unique IDs and annotations. Use when documenting project contents, creating file inventories, tracking conversation history, or building navigable knowledge maps.
kubernetes-docs
Draft, rewrite, or audit Kubernetes and cloud-native documentation using an independently expressed interpretation of Kubernetes documentation style. Use for concepts, tasks, tutorials, references, configuration, operations, and version-sensitive guidance involving clusters, workloads, APIs, kubectl, or Kubernetes…
cloudflare-observability-review
Investigate live Cloudflare Workers observability for stellar-raven-codemode. Use when reviewing production MCP request logs, playground/demo telemetry, traces, Ray IDs, request headers, telemetry query results, eval observability, agent-run forensics, or whether Cloudflare metadata is enough without app-level…
alerting-irm
Configure Grafana Alerting, Incident Response Management (IRM), and SLOs end-to-end — provisions Grafana-managed and data-source-managed alert rules, contact points (Slack/PagerDuty/email/webhook), notification policies with hierarchical matchers, silences, mute timings, on-call schedules and escalation chains…
proofpress
Preserve or inspect decision history for Markdown and static HTML knowledge artifacts with Proofpress. Use when creating or accepting a meaningful revision to an ADR, design doc, PRD, README, strategy note, portable .md artifact, or static .html/.htm artifact in a configured repository, or when receiving a file…
backup-recovery
Set up and operate the encrypted off-site backup for the chief-of-staff live data (board, guard, config, vault). Daily AES-256-GCM snapshots are pushed to a PRIVATE GitHub repo (immutable git history); the recovery key lives in the macOS Keychain. Use when bootstrapping the app on a new machine, when you need to…