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 Zfinix/aster --skill verify-before-donegit clone --depth 1 https://github.com/Zfinix/asterWrote 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/zfinix/aster/verify-before-done)<a href="https://agentmods.dev/skills/zfinix/aster/verify-before-done"><img src="https://agentmods.dev/badge/skills/zfinix/aster/verify-before-done/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/zfinix/aster/verify-before-done"><img src="https://agentmods.dev/badge/skills/zfinix/aster/verify-before-done.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00044 | $0.00447 |
| Opus 5.5 | $0.00018 | $0.00179 |
| Sonnet 5.5 | $0.00009 | $0.00089 |
| Haiku 4.5 | $0.00004 | $0.00045 |
Grade A, and why
verify-before-done scanned grade A with 1 finding 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 1mo 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
`curl -s localhost:8082/page | grep -c "expected heading"` and What it actually says
Verify before done
- Every code edit gets a check after the last edit. Typecheck or build at
minimum (
cargo check,bunx tsc --noEmit,dart analyze), tests when they exist. A check that ran before your last edit proves nothing. - Verify behavior, not just compilation. For anything that serves or
renders, probe the running artifact:
curl -s localhost:8082/page | grep -c "expected heading"andcurl -s -o /dev/null -w "%{http_code}" localhost:8082/page. A build that passes can still render the wrong thing. - Wait by polling, never by sleeping.
for i in $(seq 1 40); do curl -s localhost:8082 >/dev/null && break; sleep 0.5; done - A red check means fix, not report. Name the root cause in one sentence, fix it, re-run the same command. Never end the turn on a failing check without saying so in the first sentence.
- Calling a failure pre-existing needs evidence. Show the same failure on untouched code, or cite what does pass ("31 tests pass; the analyzer errors are a version mismatch, present before my change").
- Re-verify the neighbors. After the fix, re-run the checks that passed earlier for behavior next to your change. Breaking the sticky header while fixing the alignment is the classic complaint; replaying an earlier check catches it.
- Report with the verdict first and the evidence inline. Write: "Fixed. 15/15 tests pass, page renders the new section." Never "should work". Either "verified by " or "unverified because ".
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.
- 1mo ago First seen · 31 lines · 44 tokens per session scan A bc138649ceb3
verify-before-done is a skill published in the GitHub repository Zfinix/aster (132 stars, last pushed today), licensed Apache-2.0. It adds 44 tokens to every session and 447 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
create-aoh-pack
Use when creating or revising reusable Agentic Ops Harness packs, ops skills, workflows, agent roles, model profiles, runtime requirements, evals, or runtime adapter mappings.
speccrew-test-case-design
Designs structured test cases from Feature Spec and API Contract documents. Focuses on comprehensive test scenario analysis, test case matrix generation, and coverage traceability without involving any code implementation.
speccrew-test-code-gen
Generates executable test code from confirmed test case documents. Reads test case matrix, platform technical conventions, and system design to produce well-structured test files with full traceability to test case IDs.
speccrew-test-manager-orchestration
A test-management workflow that coordinates test design, test-code creation, test execution, and reporting.
speccrew-test-runner
SpecCrew Test Runner. Executes test code, parses results, and detects deviations between expected and actual outcomes. Reads test code files and system design to run platform-specific test suites.
speccrew-deploy-smoke-test
Performs lightweight smoke testing against running application. Verifies core API endpoint reachability based on API Contract documents. Does NOT test business logic — only HTTP status code verification.