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 skills add taniwhaai/arai --skill verifiergit clone --depth 1 https://github.com/taniwhaai/araiWrote 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/taniwhaai/arai/verifier)<a href="https://agentmods.dev/skills/taniwhaai/arai/verifier"><img src="https://agentmods.dev/badge/skills/taniwhaai/arai/verifier.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 5.1 | $0.00122 | $0.03428 |
| Opus 5 | $0.00061 | $0.01714 |
| Sonnet 5 | $0.00024 | $0.00686 |
| Haiku 4.5 | $0.00012 | $0.00343 |
Grade A, and why
verifier 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 8d 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 — 173 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verifier
Verify that an implementation or composition satisfies its contract's acceptance criteria. You did not write this code. Your job is to read the contract independently, write your own tests against the acceptance criteria, run them against the implementation, and report what passed and what failed.
Why this skill exists
The implementor wrote the code and wrote tests for it. Both came from the same context, with the same assumptions, and the same potential blind spots. If the implementor misread the contract, the implementor's tests will misread it the same way, and they will pass. "Tests pass" therefore is not "contract satisfied" — it's "implementor's interpretation of the contract is internally consistent".
Verification breaks that loop by introducing a second interpreter. You read the contract from scratch. You write tests against acceptance criteria as you understand them, not as the implementor did. You run those tests against the existing source code. If your independent reading and the implementor's converge on the same behaviour, the contract is genuinely satisfied. If they diverge, that's a finding — possibly a bug in the implementation, possibly an ambiguity in the contract, possibly a misreading on either side.
This is also why the verifier role is mandatory and not optional. An orchestrator that asks the user to "skip verification" is removing the only structural check on contract satisfaction. The skip path doesn't exist; the only options are verify-now or verify-later (deferred verification with a clearly recorded debt).
What you have
You have been given exactly four things:
- The contract — the manifest the implementation claims to satisfy. This includes the acceptance criteria, which are the load-bearing part for your work.
- The shared vocabulary — data shapes and external systems the contract refers to.
- The project context — language, toolchain commands (test/build/format/lint), code style. Use
project_context.toolchain.commands.testto run tests; never invent or re-derive language-specific test commands. - The implementation manifest — gives you the
source_pathsof the files to verify, plus any notes the implementor left.
What ships with it
4 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.
- 8d ago First seen · 173 lines · 122 tokens per session scan A e9589c7c077f
verifier is a skill published in the GitHub repository taniwhaai/arai (7 stars, last pushed 25d ago), licensed Apache-2.0. It adds 122 tokens to every session and 3,428 once invoked, about $0.0006 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
michel-create-packmind-dataset
Seed a local Packmind instance with a realistic dataset — one organization populated with standards, commands, and skills — so an autonomous agent can exercise its own changes against lifelike data instead of an empty app. Use this whenever you need populated Packmind data to verify a change end-to-end: reproducing a…
argot-check
Score your working changes with argot — flag code foreign to this repo's own patterns (unfamiliar dependencies, APIs, constructs), functions the repo already has, code filed in the wrong place, imports that break the repo's layering, and tests weakened, disabled, or deleted alongside a production change — before…
ralphctl-test-driven-development
Execute-phase skill — write the failing test before the code that makes it pass; for bug fixes, this is the reproduction test itself. Use for any logic change, bug fix, or behavioural modification; for the full root-cause triage pipeline around an unexpected failure, see ralphctl-debugging-and-error-recovery.
layer-check
Three-Layer Enforcement - Production Reality Gate.
gate-keeper
Rigorous quality guardian with auto-fix capability.
anvil
/anvil - The Anvil Quality Gate.