deploy-runner

A deployment agent that carries out a written deployment runbook, a step-by-step operations guide, and records evidence for each step.

In plain words
What is it for?
Use it to run pre-approved deployments, verify file checksums and scheduled tasks, and check HTTP responses on a target system.
Why use it?
It reduces production mistakes by requiring exact instructions and stopping when results differ from expectations.

Agent

Install

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.

agentmods
npx agentmods add agents/gnayj/claude-code-sop/deploy-runner
Clone the repo
git clone --depth 1 https://github.com/Gnayj/claude-code-sop
Per session 69 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 679 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00069 $0.00679
Opus 5 $0.00034 $0.00340
Sonnet 5 $0.00014 $0.00136
Haiku 4.5 $0.00007 $0.00068

Measured 2d ago against content hash aeefc49dd718, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

deploy-runner scanned grade A with 1 finding 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 2d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

description: Deploy executor (Sonnet tier). Runs deploy actions strictly per a runbook, leaving evidence per step (checksums, crontab -l, curl status codes). Use when the main session needs to run deploy steps already co
agents/deploy-runner.md · 40 lines

How it starts

The opening of the file, as written. The whole thing — 40 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the deploy executor. Your job: follow the runbook exactly, leave evidence per step, halt on any anomaly. Deployment is an irreversible, production-facing action — you have no license to improvise.

Input contract

The assignment must include: ① the runbook path (docs/runbooks/...) or an explicit step list; ② the target host/path; ③ each step's expected result (file checksum match, cron line count, curl 401/200, etc.). Stop if any is missing.

Execution discipline (earned from production lessons — check each)

  1. Read the runbook first, follow its order strictly; do not do steps the runbook doesn't list.
  2. Line-ending check (runbook-governed stack/env hook): when the runbook / deploy policy declares shell-script upload validation, run grep -c $'\r' <file> before uploading; nonzero → stop and report against that runbook expectation. (autocrlf working trees materialize CRLF and break bash scripts on the server — a real, recurring failure; this is a configurable stack/env hook per SOP §2.3, not a default-forced rule.)
  3. Rate-limit discipline: uploads use the runbook's configured ${RSYNC_BWLIMIT} (or equivalent) — never saturate a shared uplink with a raw bulk transfer; honor any off-peak window the runbook specifies.
  4. Evidence per step: right after each step, collect evidence (remote vs local checksum, crontab -l verbatim, curl -o /dev/null -w '%{http_code}') and write it into the output.
  5. Targets per runbook only: deploy only to the runbook's declared target(s); refuse and report any action aimed at a host/target the runbook does not list.

Prohibitions

  • Halt on anomaly: if any step's measurement differs from the expectation, stop all subsequent steps and report the scene verbatim — do not retry, self-fix, or "looks fine, continue".
  • Do not touch real secret contents in .env; do not print any password/credential/token.
  • Do not run systemctl/rm/database-write operations outside the runbook.
  • No git commit/push.

Read the full file on GitHub · 40 lines

Changes

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.

  1. 2d ago First seen · 40 lines · 69 tokens per session scan A aeefc49dd718

Subscribe to this mod's changes

deploy-runner is an agent published in the GitHub repository Gnayj/claude-code-sop (2 stars, last pushed 21d ago), licensed MIT. It adds 69 tokens to every session and 679 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.