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/arcblock/agent-skills/fleet-reportnpx skills add ArcBlock/agent-skills --skill fleet-reportgit clone --depth 1 https://github.com/ArcBlock/agent-skillsWrote 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/arcblock/agent-skills/fleet-report)<a href="https://agentmods.dev/skills/arcblock/agent-skills/fleet-report"><img src="https://agentmods.dev/badge/skills/arcblock/agent-skills/fleet-report.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 | $0.00089 | $0.01072 |
| Opus 5 | $0.00044 | $0.00536 |
| Sonnet 5 | $0.00018 | $0.00214 |
| Haiku 4.5 | $0.00009 | $0.00107 |
Grade A, and why
fleet-report scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Fleet Report — read the fleet's telemetry back
The fleet writes one JSON record per round to fleet.jsonl. This skill turns that into
numbers, then says which of them deserve attention.
The one rule
You never compute, estimate, or restate a number that the script did not print.
The script is the fact layer; you are the judgement layer. Every count, rate, percentile and duration comes from its output verbatim. A number you produce yourself is indistinguishable from a correct one to the reader — that is precisely why it is forbidden. If a number you want is not in the output, say it is not measured rather than deriving it.
Step 1 — run it
bun "${AGENTLOOP_ROOT:-$HOME/.claude/plugins/marketplaces/arcblock-agent-skills/plugins/agentloop}/scripts/fleet-report.ts" $ARGS
$ARGS from what was asked:
| Asked for | Args |
|---|---|
| default / "how's the fleet" | --days 7 |
| a specific window | --days N |
| everything | (none) |
| a dashboard / "看图" / visual | --days 7 --html /tmp/fleet-report.html then open /tmp/fleet-report.html |
| a non-default deployment | add --file <path>/fleet.jsonl |
Default to --days 7. An all-time window buries a regression that started yesterday under
weeks of healthy history.
If it exits non-zero, report the message as-is — usually the file does not exist (no fleet on
this machine, or a different logDir). Do not invent a substitute path beyond the default.
Step 2 — read the output for what matters
Report the headline numbers, then work down this list. Only raise an item if its condition is met; a healthy fleet should produce a short report.
| Look at | Raise it when | Why it matters |
|---|---|---|
| field coverage | coverage < executed | Check this FIRST. Old records lack newer fields. Every rate below is over the covered subset only — say so before quoting any of them, or you will report a young field as a fleet-wide finding. |
| residual processes | any round > 0 | A round left processes running. These accumulate and burn the machine unattended; they are invisible without this number. |
| failures | failed > 0 | Name the repo×skill and the outcome (checkout-failed ≠ setup-failed ≠ failed). |
| skipped-locked | more than an occasional one | A round collided with its own previous round still running. Means the effective cadence is longer than configured — the repo's rounds outlast its interval. |
| noop rate | high, or rising | Rounds that ran and deliberately did nothing. Healthy for a review-shaped skill (pr-sweep reviews; it does not open PRs), suspicious for a work-shaped one (issue-sweep). Judge per skill, never in aggregate. |
| P90 vs median | P90 ≫ median | A long tail. Combined with skipped-locked, that tail is what is eating the cadence. |
| produced summaries | always worth reading | The only place a round says what it did in its own words. Quote one or two verbatim rather than paraphrasing — the wording carries the reason. |
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.
- 4d ago First seen · 76 lines · 89 tokens per session scan A 4184e962dc06
fleet-report is a skill published in the GitHub repository ArcBlock/agent-skills (5 stars, last pushed 3d ago), licensed MIT. It adds 89 tokens to every session and 1,072 once invoked, about $0.0004 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
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…