Borrowing it
Nothing to install: this file belongs to bex-co/bex. 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/bex-co/bex/main/.claude/skills/loopx/SKILL.mdgit clone --depth 1 https://github.com/bex-co/bexWrote 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/bex-co/bex/loopx)<a href="https://agentmods.dev/skills/bex-co/bex/loopx"><img src="https://agentmods.dev/badge/skills/bex-co/bex/loopx/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/bex-co/bex/loopx"><img src="https://agentmods.dev/badge/skills/bex-co/bex/loopx.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 | $0.00099 | $0.04744 |
| Opus 5.5 | $0.00040 | $0.01898 |
| Sonnet 5.5 | $0.00020 | $0.00949 |
| Haiku 4.5 | $0.00010 | $0.00474 |
Grade A, and why
loopx 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 2d 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 — 171 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Task: Drain a .pm workstream item by item
/loopx <wN> — repeatedly pick the next open item in workstream <wN> — a pending milestone wN/mN/ or an open inbox note wN/NNN.md (a sub-hour task) — triage it, then give it one of four outcomes: work it and move it to done/, close it as already met, move it to blocked/, or delete it as invalid. Continue until the workstream has no open items left. This is a long-running autonomous loop over the .pm board; it composes /pm (board reads/writes), your own implementation work, and /ship.
Parse the target workstream from $ARGUMENTS (e.g. w1). If $ARGUMENTS is empty, STOP and ask which workstream to drain — never guess.
Preconditions (verify once, up front)
git branch --show-currentismain. If not, STOP and ask (same rule as/ship).git status— note pre-existing uncommitted changes, then keep going. Do not sweep unrelated changes into an item's ship: stage files by name, nevergit add -A.- Uncommitted
.pmboard work is never a reason to stop. Teammate sessions file milestones and notes into the same workstream continuously, often mid-run; that is a queue refresh, not a dirty tree. Pick the new items up in the next queue refresh and drain them like any other. If a teammate's filing is still uncommitted when you land your own outcome, commit it as its owndocs(pm): file <items>commit first — separate from yours, so authorship stays legible — then ship your work. Never edit the content of an item you did not file except to record your own verdict on it. - You share one working tree with those sessions. Before each board write, re-read the file you are about to edit (it may have changed since you last read it) and apply the smallest edit that lands your verdict. That is the whole mitigation — it does not require coordination, a pause, or a question.
- Only genuinely foreign source changes — uncommitted edits under
lego/,dashboard/,infra/,scripts/that no one in this session made and no.pmitem explains — are worth surfacing. Say what they are, leave them unstaged, and drain around them; stop only if they make the tree unshippable (see Exit).
- Uncommitted
- The workstream
.pm/<wN>/README.mdexists. If not, STOP and report. - Read
.pm/DO_NOT_DO.mdonce — it governs every triage verdict below.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 2d ago Changed · +4 lines c41bce9381f9
- 8d ago Changed · +13 lines f0a30b41b1df
- 15d ago Changed · +1 lines 90fed2b11af9
- 27d ago Changed 8b421713b976
- 1mo ago Changed · +21 lines 18079f7a2903
- 1mo ago Changed 953540098d60
- 1mo ago Changed · +16 lines 0266b2b92007
- 1mo ago First seen · 116 lines · 99 tokens per session scan A 75d4828df1c3
loopx is a skill published in the GitHub repository bex-co/bex (574 stars, last pushed today), licensed Apache-2.0. It adds 99 tokens to every session and 4,744 once invoked, about $0.0004 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 2026-09-17.
Other skills, from other repositories
kfl
KFL2 (Kubeshark Filter Language) reference. This skill MUST be loaded before writing, constructing, or suggesting any KFL filter expression. KFL is statically typed — incorrect field names or syntax will fail silently or error. Do not guess at KFL syntax without this skill loaded. Trigger on any mention of KFL, CEL…
machinist
Use Machinist to create, assign, and monitor software tasks. Use when a coding agent needs to work with Machinist, its GitHub issue workflow, intake labels, direct runs, or managed queue.
pad
Work with the user's Pad workspaces through the Pad tools. Find, read, create and update tasks, bugs, ideas, plans and docs; comment on them, link them, and see what to work on next. Use it whenever the user talks about their Pad projects or items.
product
Apply product thinking to platform work — DevEx audits, friction analysis, RFC/ADR drafting, incident communication, post-mortems, capacity planning, cost optimisation, and platform health review.
adr-workflow
Write, update or supersede an Architecture Decision Record (MADR). Load when a change picks one design among credible alternatives, sets a rule tooling or reviewers enforce, reverses an earlier ADR, or when asked to document a decision.
start-ticket
Creates the git branch for a ticket. Use when the user starts a ticket and gives its key and a short slug.