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 Intelligent-Internet/zenith --skill benchmark-validatorgit clone --depth 1 https://github.com/Intelligent-Internet/zenithWrote 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/intelligent-internet/zenith/benchmark-validator)<a href="https://agentmods.dev/skills/intelligent-internet/zenith/benchmark-validator"><img src="https://agentmods.dev/badge/skills/intelligent-internet/zenith/benchmark-validator/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/intelligent-internet/zenith/benchmark-validator"><img src="https://agentmods.dev/badge/skills/intelligent-internet/zenith/benchmark-validator.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 | $0.01454 |
| Opus 5.5 | $0.00018 | $0.00582 |
| Sonnet 5.5 | $0.00009 | $0.00291 |
| Haiku 4.5 | $0.00004 | $0.00145 |
Grade A, and why
benchmark-validator 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 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.
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 — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Benchmark Validator
Use this skill for a validation assignment targeting exactly one benchmark-related assertion.
For optimization EXP-* targets, you are not selecting the winner or optimizing
further. You independently decide whether the experiment produced a trustworthy
outcome under its contract.
For optimization EXP-* targets, act as an adversarial tester for the selected
candidate. The contract is the minimum bar, not the whole test plan. Before
promotion, define and run a compact but comprehensive validation plan that
attacks the candidate's likely correctness and performance failure modes.
For engineering VAL-* or legacy engineering targets, do not use optimization
outcome semantics. Pass only when the assigned benchmark, performance,
correctness, and guardrail behavior required by the assignment and contract is
proven with fresh evidence.
Inputs
Read:
- Validation assignment.
- The single assigned benchmark-related contract.
AGENTS.md.- Experiment ledger path cited by the assignment or contract.
- Candidate artifact/ref/patch/checksum cited by the assignment, contract, or ledger.
- Measurement protocol, correctness/guardrail commands, protected files, and source/baseline refs cited by the contract.
- For optimization targets, also inspect visible runtime resources read-only when useful: changed files, benchmark scripts, fixtures, public outputs, baseline/reference artifacts, verifier-adjacent code, logs, and generated artifacts. Do not mutate these resources, copy protected artifacts into a submission, or make submitted code depend on evaluator-only locations.
If the assignment targets more than one benchmark-related assertion, fail the assignment as too broad and request attention.
Procedure
- Identify artifacts
- Parent/baseline ref.
- Candidate ref or patch artifact.
- Benchmark command, run count, aggregation, variance policy.
- Correctness and guardrail commands.
- Protected benchmark/scoring/data/verifier files.
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 · 153 lines · 44 tokens per session scan A bf618457802d
benchmark-validator is a skill published in the GitHub repository Intelligent-Internet/zenith (372 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 44 tokens to every session and 1,454 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
rpa-init
Run when the user invokes /rpa-init or asks to onboard or warm up context on a repository. The agent studies code, reads documentation and test code, sets up the dev environment as the project expects, runs tests, and writes a short project report. No extra user brief is required.
qa
Regression test a San feature by name. Looks for a feature doc in docs/reference/ or docs/packages/2-feature/, runs automated Go tests and interactive tmux tests, then produces a pass/fail report. Use this skill when the user says "qa", "regression test", "test feature X", "verify feature", or references a feature…
rpa-gen-rules
Use when the user invokes /rpa-gen-rules or asks to create, refresh, audit, or synchronize project instructions such as root AGENTS.md, Cursor .cursor/rules, or Claude Code .claude/rules, especially when existing rule trees have drifted.
rpa-feat
Run when the user invokes /rpa-feat together with a clear feature description (for example issue text). BDD workflow: plan, failing tests, implementation, green tests, full suite, docs and examples, linter at the end. Will not work correctly without the user supplying what to build.
rpa-bugfix
Run when the user invokes /rpa-bugfix together with a clear bug description (expected vs actual, reproduction). Reproduction test first, then fix, full test suite, short report. Will not work correctly without the bug details.
ahk-test
Design, write, and run behavior-focused tests for an objective or existing code. Writes test files only and reports evidence. No tasks created, no harness tracking.