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 a1024053774/design-integrity-guardrails --skill evidence-first-testinggit clone --depth 1 https://github.com/a1024053774/design-integrity-guardrailsWrote 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/a1024053774/design-integrity-guardrails/evidence-first-testing)<a href="https://agentmods.dev/skills/a1024053774/design-integrity-guardrails/evidence-first-testing"><img src="https://agentmods.dev/badge/skills/a1024053774/design-integrity-guardrails/evidence-first-testing/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/a1024053774/design-integrity-guardrails/evidence-first-testing"><img src="https://agentmods.dev/badge/skills/a1024053774/design-integrity-guardrails/evidence-first-testing.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 | $0.00056 | $0.00945 |
| Opus 5 | $0.00028 | $0.00473 |
| Sonnet 5 | $0.00011 | $0.00189 |
| Haiku 4.5 | $0.00006 | $0.00094 |
Grade A, and why
evidence-first-testing 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 — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Evidence-First Testing
A green test proves only that the current code satisfies that test. To call a test regression evidence, show the counterfactual: the focused signal rejects the faulty behavior and accepts the repaired behavior.
Choose the claim and evidence
First classify what the red state means:
- Defect reproduction: the current implementation violates an existing contract.
- Contract test: requested new or changed behavior is not implemented yet; an expected red state is not proof of a pre-existing defect.
- Flaky or environment-specific failure: logs, inputs, seeds, timing, versions, and environment details must travel with the result.
Before implementing or repairing the claimed behavior, obtain the smallest reliable signal:
- an existing failing test;
- a new focused test or minimal reproducer;
- a build, type-check, request, trace, or command that exposes the defect;
- for flaky or environment-specific failures, preserved logs, inputs, seeds, timing, and environment details.
Record the exact command or procedure, inputs, expected result, observed result, environment, and exit status. Confirm that the signal fails for the expected reason. An unrelated setup, fixture, import, or environment failure is not red-state evidence.
Enforce the gate
For bug fixes, regression tests, and behavior changes with a testable contract, use this order:
- Inspect and diagnose without changing the relevant behavior.
- Run the focused signal against the current code and record the red-state result.
- Make the smallest justified implementation change.
- Re-run the same signal and confirm it turns green.
- Run the smallest broader suite or check that is required by the repository, the user, or a still-credible failure category. Broaden only when the changed contract, boundary, or failure mode requires it; do not use a fixed test count or line-count comparison as a substitute for evidence.
Do not begin the implementation step until a focused signal exists. A bug fix requires defect-specific red-state evidence; a new behavior may use an expected-red contract test. If no safe red state or counterfactual can be established, report the evidence as INCOMPLETE instead of calling the test regression-proof. Reading code, adding diagnostic logging, or creating a test fixture is allowed when it does not repair or bypass the claim.
What ships with it
1 file 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 · 95 lines · 56 tokens per session scan A 2e3fcd8850a0
evidence-first-testing is a skill published in the GitHub repository a1024053774/design-integrity-guardrails (0 stars, last pushed today), licensed MIT. It adds 56 tokens to every session and 945 once invoked, about $0.0003 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 2026-09-07.
Other skills, from other repositories
gentle-ai-bench
Trigger: bench, journey, journeys, driven mode, gentle-ai-bench, journey corpus, j-numbers, bench axis. Author and verify gentle-ai bench journeys; go test ./bench never proves driven execution.
go-testing
Trigger: Go tests, go test coverage, Bubbletea teatest, golden files. Apply focused Go testing patterns.
gentle-ai
Use Gentle AI harness discipline for Pi work: clarify first, preserve OpenSpec artifacts, use strict TDD where available, delegate through subagents when useful, and protect review workload.
include-test-files-that-assert-on-behavior-being-changed-in-decl
When delegating a task affected by this skill, include.
qa
Use to verify that code works correctly — browser-based testing with Playwright, native app testing with computer use, CLI testing, API testing, or root-cause debugging. Supports --quick, --standard, --thorough modes. Triggers on /qa.
ring:adding-multi-tenancy
Adding database-per-tenant isolation into a Go service end-to-end via an 11-gate cycle: detects the stack, audits compliance, then dispatches backend agents to implement tenantId-from-JWT routing through the lib-commons v5 dispatch layer (config, middleware, repositories, metrics, tests) and runs reviewers. Use when…