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 shenli/devtool-ax-kit --skill ax-verificationgit clone --depth 1 https://github.com/shenli/devtool-ax-kitWrote 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/shenli/devtool-ax-kit/ax-verification)<a href="https://agentmods.dev/skills/shenli/devtool-ax-kit/ax-verification"><img src="https://agentmods.dev/badge/skills/shenli/devtool-ax-kit/ax-verification/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/skills/shenli/devtool-ax-kit/ax-verification"><img src="https://agentmods.dev/badge/skills/shenli/devtool-ax-kit/ax-verification.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 | $0.00252 |
| Opus 5.5 | $0.00006 | $0.00101 |
| Sonnet 5.5 | $0.00003 | $0.00050 |
| Haiku 4.5 | $0.00002 | $0.00025 |
Grade A, and why
ax-verification 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 14d 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.
What it actually says
Define assertions outside the agent's editable workspace. Verify observable behavior, not implementation shape: functional outputs, persistence across a restart, independent identities, effect counts, and credential absence where relevant. Return structured pass/fail results with details and a nonzero exit status on failure.
Run the verifier separately after the agent finishes. Do not let the agent edit the verifier or turn a missing assertion into a pass. Keep checks local or explicitly scoped to disposable test resources.
For timing-sensitive behavior, deliberately exercise the boundary and use a fresh identity per trial. Report valid-trial denominators separately from infrastructure availability.
Record disagreement between the agent's reported result and the independent result as a first-class outcome. Exercise timing, restart, duplicate-effect, and malformed-response boundaries when those are part of the task contract.
For CLI surfaces, capture stdout, stderr, and exit status separately. Validate
the promised data format without cleaning the output first, and compare file
inventories before and after read operations. Use docs/cli-evaluation.md for
case selection and distinguish a captured child's failure from wrapper misuse.
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.
- 14d ago Changed · +5 lines d70e0a4eb89a
- 2mo ago First seen · 23 lines · 15 tokens per session scan A ce0c783ff33b
ax-verification is a skill published in the GitHub repository shenli/devtool-ax-kit (2 stars, last pushed 16d ago), licensed Apache-2.0. It adds 15 tokens to every session and 252 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 skills, from other repositories
using-axis
Run AXIS, read its reports, navigate its project layout, and interpret scores. Use when the user asks to run AXIS, invoke the CLI, run a specific profile or suite, compare runs, explain a score, find a regression, manage baselines, or understand where AXIS writes its files.
mandu-mcp-verify
An automatic verification workflow for the Mandu project that runs after edits, generated files, or before a commit. It checks generated behavior, architecture rules, code quality, and project structure in parallel.
mandu-testing
Testing patterns for Mandu applications. Use when writing unit tests, integration tests, or E2E tests. Triggers on test, spec, Bun test, Playwright, or testing tasks.
configure-axis
Author AXIS (Agent Experience Index Score) scenarios and axis.config.json for a project. Use when the user asks to set up AXIS, add a scenario, write or edit axis.config.json, define profiles or split scenarios into suites, stage an external repo into a scenario workspace, or evaluate an AI agent with AXIS.
TDD Methodology
This skill should be used when the user asks to "write tests first", "use TDD", "test-driven development", "red green refactor", "test first", "add unit tests before code", "write regression test first", "safe refactor with tests", or when TDD mode is active and the user makes any coding request that affects behavior…
cf-fix
Use when asked to FIX a bug, error, test failure, CI/CD issue, type error, lint error, log error, UI issue, or code problem. Uses cf-debug for evidence-first diagnosis before any code change. Use when: Invoke to fix a bug or failure with scout-first diagnosis before change.