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 agentmods add skills/jeonje/claude-plugins/final-checknpx skills add JeonJe/claude-plugins --skill final-checkgit clone --depth 1 https://github.com/JeonJe/claude-pluginsWrote 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/jeonje/claude-plugins/final-check)<a href="https://agentmods.dev/skills/jeonje/claude-plugins/final-check"><img src="https://agentmods.dev/badge/skills/jeonje/claude-plugins/final-check.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 | $0.00118 | $0.01569 |
| Opus 5 | $0.00059 | $0.00785 |
| Sonnet 5 | $0.00024 | $0.00314 |
| Haiku 4.5 | $0.00012 | $0.00157 |
Grade A, and why
final-check 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 — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Final Check
Deliverable-side quality gate, run after implementation and review are done and before the work leaves your machine. The reader on the other side has none of your context — this gate verifies the artifact stands alone.
Every check produces evidence (command output, file citation). A check without evidence is ❓, never ✅.
Phase 1 — Requirements coverage map
Take the original requirement source (spec document, issue, request message) and build the coverage table from the requirements side, not from what you built:
| # | Stated requirement | Where implemented (file/symbol) | Where verified (test/manual evidence) | Status |
|---|
- Status:
COVERED/PARTIAL(state what's missing) /NOT COVERED(state why — deliberate descope needs one line of reasoning the reader will see) /EXTRA(built but never asked — flag it; unrequested scope reads as not understanding the ask). - Ambiguous requirements you interpreted: record the interpretation and where it's documented for the reader.
- No requirement source available → reconstruct the implied checklist from the deliverable's own README/claims, and verify the artifact delivers what it promises.
Phase 2 — Clean-clone reproducibility
The single highest-failure gate. Verify in an isolated copy, not your working tree:
git clone <repo> /tmp/final-check-clone && cd /tmp/final-check-clone
# then follow README literally, top to bottom
- Execute the README's setup/run/test instructions exactly as written — no steps from memory. Every command you must run that the README doesn't mention is a README bug.
- Check declared prerequisites vs reality: language/runtime version, package manager, required env vars (is there an
.env.example?), external services (DB, docker) — can the reader actually satisfy them? - Run the test suite from the clean clone and record the fresh output. "It passed earlier in my working tree" is not evidence.
- If a step cannot be run in your environment, mark it ❓ with the reason — do not silently assume it works.
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 · 103 lines · 118 tokens per session scan A 90e0befea3ba
final-check is a skill published in the GitHub repository JeonJe/claude-plugins (3 stars, last pushed 1mo ago), licensed MIT. It adds 118 tokens to every session and 1,569 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
execution-grounded-review
Execution-grounded review: run tests first, trace each acceptance criterion to execution evidence. Use when verifying an implementation meets spec.
multi-model-delegation
Multi-model design consults via PAL (kimi, glm, gemini, gpt). Use when asking other models to brainstorm a design or reconciling their split answers.
blueprint-work-order
Create a work-order for isolated subagent execution, optionally linked to a GitHub issue. Use when breaking a PRP into delegatable tasks or spawning from an issue.
document-detection
Detect PRD/ADR/PRP opportunities in conversations and prompt for document creation. Use when the user discusses feature requirements, tech trade-offs, or implementation plans.
config-sync
Config sync across FVH repos: extract, diff, propagate tooling improvements. Use when syncing workflows or configs across multiple repos.
document-linking
Unified ID system for PRDs, ADRs, PRPs, and GitHub issues with bidirectional links. Use when linking docs, finding orphans, auto-assigning IDs, or validating cross-doc references.