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/appautomaton/automaton/auto-verifynpx skills add appautomaton/automaton --skill auto-verifygit clone --depth 1 https://github.com/appautomaton/automatonWrote 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/appautomaton/automaton/auto-verify)<a href="https://agentmods.dev/skills/appautomaton/automaton/auto-verify"><img src="https://agentmods.dev/badge/skills/appautomaton/automaton/auto-verify.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 | $0.00018 | $0.01402 |
| Opus 5 | $0.00009 | $0.00701 |
| Sonnet 5 | $0.00004 | $0.00280 |
| Haiku 4.5 | $0.00002 | $0.00140 |
Grade A, and why
auto-verify 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 5d 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 — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.
auto-verify
Verification gate. Independent audit of a completed plan. It runs once, not per-slice.
First action: run node .agent/.automaton/scripts/get-context.mjs from the project root.
Preamble
Independent audit. Re-read the plan, run proof commands, and compare fresh results to acceptance criteria. It does not trust execute's self-assessment or fix what it finds. When continuing inline from execute, re-derive from fresh command output; execute's reasoning is context, not evidence.
Loading discipline: on warm continuation from execute, one PLAN.md read plus verification commands per criterion. On cold entry, load in dependency order, spec before plan; auto-resume owns the full order table. Read source files when verifying correctness requires inspecting the actual changes, not just command output.
Quality Gate
Before writing the verification report:
- Tie every result to fresh command output or direct observation.
- Name skipped checks explicitly. Omission is not a pass.
- Treat partial evidence as FAIL for the plan.
- Read
references/quality.mdwhen the report sounds confident without proof.
Do
Load State
Read the canonical PLAN.md. Load only linked slices/slice-NNN.md files and referenced requirement IDs from spec/*.md: linked detail and traceability IDs are normative, and an unlinked supplemental file is not verification context. For prose slices, read references/content-verification.md.
Mark Verify Stage
After PLAN.md resolves and before running commands, run node .agent/.automaton/scripts/sync-status.mjs --stage verify from the project root.
Collect Acceptance Criteria
Gather every acceptance criterion and verification command from every slice in PLAN.md. Build a checklist: slice name → criterion → command. This is a plan-level audit.
Do NOT modify source code, tests, or project artifacts during verification. Verify reads and runs commands; its only writes are the markdown records this skill owns (VERIFY-GAP blocks, the ## Verification section, the ROADMAP phase update). It does not fix.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 5d ago First seen · 87 lines · 18 tokens per session scan A b9fdafefe77a
auto-verify is a skill published in the GitHub repository appautomaton/automaton (21 stars, last pushed 18d ago), licensed MIT. It adds 18 tokens to every session and 1,402 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
ahk-test
Design, write, and run behavior-focused tests for an objective or existing code. Writes test files only and reports evidence. No tasks created, no harness tracking.
ahk-review
Preview a code review against ticket/objective alignment, with deep semantic (name-vs-behavior) analysis. No tasks created, no harness tracking.
ahk-triage
Triage a bug or unexpected behavior. Deep diagnostic analysis with structured report. No tasks created, no harness tracking.
ahk-consultant
Get technical advice or review on an approach, idea, or change. References available skills for best practices. No tasks created, no harness tracking.
ahk-ask
Ask a question about this codebase — where is X, does Y exist, how does Z work. Read-only. No tasks created, no harness tracking.
assumptions
Generate an evidence-backed Assumptions ledger for a Git diff, feature, endpoint, worker, migration, or code path. Use when reviewing a change, asking what could break, preparing a release, finding edge cases, assessing deployment safety, analyzing retries or concurrency, or designing failure-focused tests.