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.
git clone --depth 1 https://github.com/tanghong123/claude-replayWrote 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/commands/tanghong123/claude-replay/monitor-fleet)<a href="https://agentmods.dev/commands/tanghong123/claude-replay/monitor-fleet"><img src="https://agentmods.dev/badge/commands/tanghong123/claude-replay/monitor-fleet/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/commands/tanghong123/claude-replay/monitor-fleet"><img src="https://agentmods.dev/badge/commands/tanghong123/claude-replay/monitor-fleet.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 | $0.00024 | $0.00303 |
| Opus 5 | $0.00012 | $0.00151 |
| Sonnet 5 | $0.00005 | $0.00061 |
| Haiku 4.5 | $0.00002 | $0.00030 |
Grade A, and why
monitor-fleet 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 9d 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.
What it actually says
Aggregate the user's claude-monitor instances into one page with
claude-monitor-fleet.
Do it in this order, and do not invent host names or ports — everything comes
from discovery or from the user. $ARGUMENTS, if given, is an extra SSH
destination to include.
-
Survey what actually exists (this writes nothing):
claude-monitor-fleet discoverWith an argument:
claude-monitor-fleet discover --host $ARGUMENTS -
Show the user the list. If it is empty, say so — no monitor is running anywhere reachable — and stop rather than guessing a port.
-
Save it, then serve it:
claude-monitor-fleet discover --add claude-monitor-fleet up
up holds the tunnels open in the foreground and they close when it exits, so
run it where it will outlive this turn and tell the user the
http://127.0.0.1:<port>/ URL it prints. A tunnel that drops later is re-opened
by that same process, with backoff — a red dot is not a reason to restart it.
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.
- 9d ago First seen · 36 lines · 24 tokens per session scan A 6047645ed71c
monitor-fleet is a command published in the GitHub repository tanghong123/claude-replay (2 stars, last pushed yesterday), licensed MIT. It adds 24 tokens to every session and 303 once invoked, about $0.0001 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 commands, from other repositories
nab
Fetch, archive, or research URLs with nab.
check
Audit a .nika.yaml workflow before it runs — findings with their NIKA-XXXX codes, cost envelope, permits.
trace
Read a run's flight recorder — verdict, failing task, tamper check, and the exact resume line if paused.
doctor
Diagnose this machine's Nika surface — binary, plugin kits, providers, wiring — every problem with its exact fix command.
explain
Narrate a workflow (waves · cost · touches · how to run) or teach a NIKA-XXXX error code.
dangerous-deploy
Command "dangerous-deploy" from ryo-ebata/cc-audit, covering deploy command, steps, build, deploy with elevated privileges and sync secrets to remote.