Borrowing it
Nothing to install: this file belongs to epam/ai-dial-chat. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/epam/ai-dial-chat/development/.claude/skills/lean-verification/SKILL.mdgit clone --depth 1 https://github.com/epam/ai-dial-chatWrote 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/epam/ai-dial-chat/lean-verification)<a href="https://agentmods.dev/skills/epam/ai-dial-chat/lean-verification"><img src="https://agentmods.dev/badge/skills/epam/ai-dial-chat/lean-verification.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00040 | $0.00491 |
| Opus 5 | $0.00020 | $0.00246 |
| Sonnet 5 | $0.00008 | $0.00098 |
| Haiku 4.5 | $0.00004 | $0.00049 |
Grade A, and why
lean-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 4d 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 — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Lean verification
Use the narrowest verification tier that can disprove the current change. The
scripts capture full output under tmp/agent-logs/; successful runs return a
two-line status, while failures return a bounded excerpt and the log path.
Verification tiers
# Red/green loop: pass one or more workspace-relative test paths.
npm run test:file -- libs/example/src/Foo.spec.ts
# Completed task or vertical slice: affected projects only.
npm run verify:changed
# Final pre-merge gate: full, non-mutating verification.
npm run verify:full
test:fileis the default during implementation. It groups paths by their nearest Vite/Vitest config and runs only those files.test:changedandverify:changeduse Nx's affected project graph withorigin/developmentas the base.verify:affectedremains available as an explicit affected verification alias.verify:fulluseslint:check, never the mutatingnpm run lintcommand.build:quietbuilds affected projects when the change can affect bundling.
This workspace contains many Nx projects, so affected verification narrows typecheck, lint, test, and build work to the changed projects and their affected dependents. Prefer exact tests for the mid-task loop; reserve project-level affected checks for a completed slice. A successful affected run may execute no task when only docs or agent configuration changed, so it is not a substitute for the final full gate.
Verification cadence
Batch related edits for one task-list item before re-verifying. Do not rerun a suite after each individual edit. OpenSpec tasks should name their exact test files so the red/green loop does not require rediscovery.
Failure handling
Use the printed failure excerpt first. Search or read only the relevant portion
of the saved log if that excerpt is insufficient; do not load the whole log or
rerun with a verbose reporter by default. Keep --reporter=minimal so passing
test logs do not enter the conversation.
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.
- 4d ago First seen · 54 lines · 40 tokens per session scan A ef75ffea76c4
lean-verification is a skill published in the GitHub repository epam/ai-dial-chat (504 stars, last pushed today), licensed Apache-2.0. It adds 40 tokens to every session and 491 once invoked, about $0.0002 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 2026-09-03.
Other skills, from other repositories
browser-use
Direct browser control via CDP for web interaction: automation, scraping, testing, screenshots, and site/app work.
mantis-configure
Configures and validates Mantis pipeline environments, sandbox mechanisms, and AI models. Use to set up workflow.json, auto-detect host capabilities, switch between sandboxes (static-only, gvisor, microsandbox, gce), select AI models, and run fast 1-2s preflight tests. Don't use for scanning source code or running…
output-dev-eval-testing
Create offline evaluation tests for Output SDK workflows using @outputai/evals. Use when implementing test evaluators with verify(), creating dataset YAML files, building eval workflows, or running workflow tests via CLI.
output-dev-scenario-file
Create test scenario JSON files for Output SDK workflows. Use when creating test inputs, documenting expected behaviors, or setting up workflow testing.
output-plan-workflow
Use when the user asks to create, build, generate, scaffold, or plan a new workflow. Orchestrates the full planning process including architecture, steps, prompts, evaluators, and testing strategy using specialized subagents.
copywriting
Direct-response copywriting for landing pages, emails, and ads. Picks the right framework (AIDA, PAS, 4U headlines), loads brand voice if a profile exists, builds a 10-variant headline matrix, writes proof-backed body copy toward a single CTA, and pressure-tests every line. Use when the user says "write the landing…