Borrowing it
Nothing to install: this file belongs to haoyifan/Silicon-Pantheon. 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/haoyifan/Silicon-Pantheon/master/.claude/skills/review-system-test/SKILL.mdgit clone --depth 1 https://github.com/haoyifan/Silicon-PantheonWrote 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/haoyifan/silicon-pantheon/review-system-test)<a href="https://agentmods.dev/skills/haoyifan/silicon-pantheon/review-system-test"><img src="https://agentmods.dev/badge/skills/haoyifan/silicon-pantheon/review-system-test/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/haoyifan/silicon-pantheon/review-system-test"><img src="https://agentmods.dev/badge/skills/haoyifan/silicon-pantheon/review-system-test.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 | $0.00000 | $0.01746 |
| Opus 5.5 | $0.00000 | $0.00698 |
| Sonnet 5.5 | $0.00000 | $0.00349 |
| Haiku 4.5 | $0.00000 | $0.00175 |
Grade A, and why
review-system-test 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 2mo 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 — 191 lines — stays where its author put it; the contents beside it link to each section on GitHub.
System-Test Bundle Review Skill
You are an experienced SRE + backend engineer reviewing a
system-test bundle produced by silicon-system-test. The bundle
captures everything from one run: server logs, per-agent logs,
replay files, plus the orchestrator's own log and manifest.
Input
`` is the path to a bundle directory. If empty, look under
~/silicon-system-test-results/ for the most recent one:
ls -t ~/silicon-system-test-results/ | head -1
Layout you should expect:
/
run-manifest.json # machine-readable: agents, config, outcomes
INCIDENTS.md # orchestrator-detected problems (pre-written)
orchestrator.log # what the orchestrator did, when
server/
silicon-serve.stdout.log
*.log # silicon-serve's log file (pid-and-ts named)
replays/*.jsonl # one replay per completed match
leaderboard.db # sqlite, if any matches counted
clients/
-host.toml
-host.log
-host.stdout.log
-joiner.toml
...
What to check
Walk through the list, report every issue with severity: CRITICAL (run can't be trusted), HIGH (likely regression), MEDIUM (flaky / warning), LOW (noise), INFO.
1. Manifest summary (CRITICAL if missing)
Read run-manifest.json:
summary.n_crashed > 0→ each crashed agent is a HIGH finding; name it and cite the stdout tailsummary.n_killed_by_timeout > 0→ HIGH; the run ran out of wall clocktimed_out: true→ CRITICAL in most cases (random-vs-random should never hit the 4 h cap)config.run.num_matches vs summary.n_clean_exit— matches that didn't reach game_over are worth flagging
2. Orchestrator-detected incidents (CRITICAL-for-each)
Read INCIDENTS.md. The orchestrator has already pre-classified obvious
failures. Every line there is at minimum HIGH; treat as first-class
findings in your report.
3. Server log — crashes + invariants (CRITICAL)
server/*.log (not the stdout one — the silicon-*.log file):
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.
- 2mo ago First seen · 191 lines · 0 tokens per session scan A ca8baef97cfb
review-system-test is a skill published in the GitHub repository haoyifan/Silicon-Pantheon (7 stars, last pushed 5mo ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,746 tokens. 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
systematic-debugging
4-phase root cause debugging: understand bugs before fixing.
module-regression
Govern cross-module regression in large repositories with a REGRESSION.md ledger that records each module's downstream consumers and executable acceptance command. After a change, run the changed module and every affected downstream command, and use observed exit codes as regression evidence under the project's…
fix-failing-tests
Diagnose a failing test in the googleapis/mcp-toolbox repo and land a fix by reasoning from the actual error: read the failure, reproduce it, shrink it until the cause is forced into the open, then fix the cause. Use this whenever a test or CI job is red, a build breaks after a change, many packages fail at once, or a…
reproduce-bug
Reproduce a reported bug in googleapis/mcp-toolbox and decide whether it is real, delivering an evidence-backed verdict: confirmed, already fixed, misconfiguration, client-side, works as intended, not reproducible, or blocked. Use whenever a maintainer asks you to reproduce, verify, confirm, or investigate a bug…
credit-note-fixer
Fix the tiny credit-note formatting bug and rerun the exact targeted test command.
verify-omnigent
Drive Omnigent the way a user does and prove a behavior with recorded evidence, using an isolated server, runner, and mock model plus a feature map of every user entry point. Load before reproducing a user-facing bug, before claiming a fix works, or when reviewing whether a change covered every surface (session vs.…