Borrowing it
Nothing to install: this file belongs to juspay/neurolink. 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/juspay/neurolink/release/.claude/skills/neurolink-ci-maintenance/SKILL.mdgit clone --depth 1 https://github.com/juspay/neurolinkWrote 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/juspay/neurolink/neurolink-ci-maintenance)<a href="https://agentmods.dev/skills/juspay/neurolink/neurolink-ci-maintenance"><img src="https://agentmods.dev/badge/skills/juspay/neurolink/neurolink-ci-maintenance/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/juspay/neurolink/neurolink-ci-maintenance"><img src="https://agentmods.dev/badge/skills/juspay/neurolink/neurolink-ci-maintenance.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 | $0.00059 | $0.03488 |
| Opus 5.5 | $0.00024 | $0.01395 |
| Sonnet 5.5 | $0.00012 | $0.00698 |
| Haiku 4.5 | $0.00006 | $0.00349 |
Grade B, and why
neurolink-ci-maintenance scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Recursive force deletemediumDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
rm -rf "$(dirname "$audit_dir")" Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
How it starts
The opening of the file, as written. The whole thing — 246 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CI, generators and the advisory gate
Moved out of the root CLAUDE.md. The required-status-checks section is still in the root CLAUDE.md; stacked-PR and merge playbooks are in the neurolink-pr-merge skill.
⚠️ A "regenerate and diff" check needs a reproducible generator first
docs/api currency works because typedoc is a pure function of the source.
docs-site/static/{llms.txt,llms-full.txt,search-index.json} were not, and a
naive currency check over them would have been permanently red — the same
shape of always-failing required check the ffmpeg incident below produced.
(Only search-index.json is committed now. The two llms files are gitignored:
they are not in package.json files, so they never shipped, and committing a
9.5 MB whole-file regeneration conflicted on nearly every docs pull request.
The reproducibility lesson below still stands — it is what lets the remaining
artifact be diffed at all.)
Two separate causes, and the second is the one that hides:
build-llms-txt.tsstampednew Date().toISOString()into both llms files. Obvious once you look, and trivially fatal to any diff.docFiles.sort((a, b) => a.order - b.order)is not a total order — many files share oneorder.Array.prototype.sortis stable, so ties fell back to the input order, which came fromglob(), which is filesystem order. Two consecutive builds of an unchanged tree differed by 215,862 lines.sortSectionshad the same defect: every section outsideSECTION_ORDERscores 999, so their ties came from Map insertion order.
Both are fixed in the llms generator, and those two artifacts are now byte-identical across runs.
search-index.json was not fixed with them, and this file said otherwise for
months. The fix commit ef125b2bd touched five paths — the workflow, this
file, build-llms-txt.ts, and the two llms artifacts. grep -c search-index
over that file list returns 0. Its generator,
docs-site/plugins/docusaurus-plugin-search-index/index.js, walked
fs.readdirSync with no sort at all and numbered entries with a positional
counter (objectID: String(id++)), so any reordering rewrote every later id and
turned a one-file change into a whole-file diff; a rebuild on an unchanged tree
differed by ~1,640 positions. That was the only one of the three artifacts still
committed, and so the only one a currency check actually diffs.
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 · 246 lines · 59 tokens per session scan B 672ecc8031c5
neurolink-ci-maintenance is a skill published in the GitHub repository juspay/neurolink (168 stars, last pushed yesterday), licensed MIT. It adds 59 tokens to every session and 3,488 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it B with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-11-02.
neurolink-ci-maintenance compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| neurolink-ci-maintenancejuspay/neurolink | Skill | 168 | yesterday | MIT | B | 59 | Claude Code |
| swarm-migrateyonatangross/orchestkit | Skill | 299 | yesterday | MIT | A | 103 | Claude Code |
| ci-debugyonatangross/orchestkit | Skill | 299 | yesterday | MIT | A | 55 | Claude Code |
| devops-deploymentyonatangross/orchestkit | Skill | 299 | yesterday | MIT | A | 34 | Claude Code |
| documentation-patternsyonatangross/orchestkit | Skill | 299 | yesterday | MIT | A | 53 | Claude Code |
| zb-release-pipelineAdamBien/airails | Skill | 62 | yesterday | MIT | A | 148 | Claude Code, Codex |
| fluent-ci-cdserac-labs/serac | Skill | 77 | 17d ago | Apache-2.0 | A | 78 | Claude Code |