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.
git clone --depth 1 https://github.com/LeanAndMean/scramjetWrote 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/commands/leanandmean/scramjet/mach12-push)<a href="https://agentmods.dev/commands/leanandmean/scramjet/mach12-push"><img src="https://agentmods.dev/badge/commands/leanandmean/scramjet/mach12-push/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/leanandmean/scramjet/mach12-push"><img src="https://agentmods.dev/badge/commands/leanandmean/scramjet/mach12-push.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00015 | $0.00979 |
| Opus 5.5 | $0.00006 | $0.00392 |
| Sonnet 5.5 | $0.00003 | $0.00196 |
| Haiku 4.5 | $0.00002 | $0.00098 |
Grade A, and why
mach12:push 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 1mo 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 — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Push
Goals
- Commit and push exactly one caller-bounded batch without including unrelated work or likely secrets.
- Verify local, upstream, and forge identities converge after the push, preserving exact proof provenance in validation mode.
- Return a durable progress artifact for ordinary work whenever an associated pull request or issue can be established.
This command is delegate-only. Routing belongs to the caller.
Step 1: Establish the bounded change
Run git status, inspect staged and unstaged diffs, and use the caller context to identify exactly which files belong to this batch. Never use git add . or git add -A, and never stage likely secrets.
When the caller supplies an accepted validation-proof payload, treat it as a distinct mode. Before repository or remote mutation, require:
- an authenticated open PR, repository, head branch, upstream destination, and matching upstream repository;
- local
HEAD, upstream, and fresh GitHubheadRefOidall equal the supplied implementation parentP; - an empty index;
- worktree changes consisting exactly of the supplied accepted test paths;
- a tests-only diff with no production, temporary, unrelated, or secret-bearing content;
- every supplied node discoverable with the expected red assertion result.
Stage only the accepted test paths and require the staged diff to equal the complete worktree diff with no residual changes. If any boundary is unclear or false, stop before committing or pushing and report the observed state. Do not infer missing accepted paths or adapt the proof set.
For ordinary work, stage only the files known from the current session. If ownership is unclear, ask the user rather than guessing.
Step 2: Commit
Review recent commit messages for repository style and create one concise commit describing why the bounded change exists. Do not add model or tool co-author footers unless established repository history requires them.
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.
- 1mo ago Changed · +4 lines 46c90daaf84b
- 1mo ago First seen · 70 lines · 15 tokens per session scan A 4928405650a8
mach12:push is a command published in the GitHub repository LeanAndMean/scramjet (8 stars, last pushed 10d ago), licensed Apache-2.0. It adds 15 tokens to every session and 979 once invoked, about $0.0001 per session on Opus 5.5. 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 commands, from other repositories
afcl
Close feature [agent] [--adopt] - merges branch, cleans up, optionally adopts from losers (shortcut for feature-close).
afsr
Review feature spec - improve the spec itself before implementation (shortcut for feature-spec-review).
afrv
Revise the current feature worktree after code review — decide accept/revert/modify (shortcut for feature-code-revise).
afsrv
Revise feature spec after pending spec reviews — decide and acknowledge in one pass (shortcut for feature-spec-revise).
afp
Prioritise feature - assigns ID, moves to backlog (shortcut for feature-prioritise).
research-spec-revise
Revise research spec after pending spec reviews — decide and acknowledge in one pass.