acp

A command that stages all project changes, creates a descriptive Git commit, and pushes it to the remote repository. Git is a system for tracking code changes, while a remote repository is the shared copy stored elsewhere.

In plain words
What is it for?
Use it when you want to review current changes, commit them, and send them to the remote repository, including when working with stacked branches or commits.
Why use it?
It combines the usual save-and-share steps and checks for likely secrets before staging files. This helps prevent accidental commits of passwords, private keys, or credentials.

Command for Claude Code

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 commands/cbmono/ai-setup/acp
Clone the repo
git clone --depth 1 https://github.com/cbmono/ai-setup

Made for: Claude Code.

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 959 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 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 $0.00000 $0.00959
Opus 5 $0.00000 $0.00479
Sonnet 5 $0.00000 $0.00192
Haiku 4.5 $0.00000 $0.00096

Measured 2d ago against content hash db53764e2f36, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

acp scanned grade C 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.

Reaches for credential fileshighPrivilege escalation

SSH keys, cloud credentials, git-credentials, .npmrc, /etc/shadow: reading these is how a config file becomes a credential leak.

`id_ecdsa` / `id_dsa`; `credentials`, `service-account*.json`, `*.kubeconfig`, `.npmrc`, `.netrc`.
.claude/commands/acp.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.

Stage all changes, create a commit with a descriptive message, and push to the remote. Stack-aware.

Steps

  1. Scan for secrets BEFORE staging anything. git add -A stages every non-ignored path, so a secret staged first is a secret the rest of this flow has to un-stage. List the paths git would stage — git status --porcelain --untracked-files=all (--untracked-files=all matters: plain --porcelain collapses an untracked directory to dir/ and would hide dir/.env) — and match the paths only against: .env and .env.* except the templates .env.example / .env.template / .env.sample; *.pem, *.key, *.p12, *.pfx, *.keystore, *.jks; id_rsa / id_ed25519 / id_ecdsa / id_dsa; credentials, service-account*.json, *.kubeconfig, .npmrc, .netrc. If anything matches, stop before staging and report the matching paths and the .gitignore / git rm --cached fix. Only continue if the user tells you explicitly to include that path.
  2. git add -A to stage all changes.
  3. Run git status and git diff --cached to understand what's being committed. If step 1 matched anything the user then authorised, exclude those paths from the diffgit diff --cached -- . ':(exclude)<path>' for each — because a bare git diff --cached prints the file's contents to the terminal and into this transcript. Deciding not to read the output is not the same as not emitting it, and the standing rule is that secrets are never printed or logged, not that they are printed and ignored.
  4. Write a concise, descriptive commit message based on the changes. If $ARGUMENTS is provided, use it verbatim as the commit message instead.
  5. Commit.
  6. Decide how to push:
    • If the current branch is part of a gh stack (run gh stack view and check it doesn't error), run gh stack submit — it both pushes the stack and creates/updates the PRs in one step, so the PR list and the remote tips stay aligned. (gh stack push only pushes branches; PRs would still update via the new tip, but submit also reconciles PR titles/descriptions with stack ordering, which is what we want after a commit.)
    • Otherwise, git push on the current branch (add -u origin <branch> if the branch has no upstream).
  7. Report the commit SHA and, for stacked branches, the PR URL from gh stack view. If the diff you read in step 3 suggests a follow-up review pass would catch something — substantial logic changes, a new module, a refactor across several files in one area, a large delete — append a one-line "consider /verify" / "consider /grill" / "consider /scan on <dir>" / "consider /techdebt" suggestion tied to the actual signal. Skip the suggestion entirely for doc-only, config-only, dep-bump, or formatting commits. Don't run the suggested command — the user picks.

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 · 0 tokens per session scan C db53764e2f36

Subscribe to this mod's changes

acp is a command published in the GitHub repository cbmono/ai-setup (2 stars, last pushed 8d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 959 tokens. A static security scan graded it C with 1 finding (reaches for credential files). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.