audit-remediations
Skill2 starsChanged 2 days agoMITView on GitHub
Use when deciding which patching scripts to add next, or auditing whether existing scripts still cover the live tenant. Hits the dev server's /vulnerabilities page, extracts every recommended-remediation string, buckets them by vendor/product, and prints uncovered remediations ranked by host count. Trigger phrases: "what scripts should we add next", "audit remediations", "what's uncovered", "which patches are surfaced".
In this repository8 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- audit-write-toolsSkill
- cost-summarySkill
- dev-serverSkill
- patch-contextSkill
- validate-bundleSkill
- verify-featureSkill
- verify-patchSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 15
If the dev server isn't up, ask the user to run npm run dev (or docker compose up -d for the container path) first.
- Distinctive: line 38
If the user wants the audit but the dev server isn't reachable (remote work, broken local stack), say so explicitly.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
audit-remediations
Re-pull the live Spotlight remediation surface and diff against the patching catalog.
Prerequisites
- Dev server up at
http://localhost:3000(the/vulnerabilities?mode=newest&limit=2000route renders the Spotlight fanout). - Falcon credentials valid in the running process — without them the page renders but every row is a 403.
If the dev server isn't up, ask the user to run npm run dev (or docker compose up -d for the container path) first. Don't try to start it for them.
What to run
bash .claude/skills/audit-remediations/run.shThe script:
curls the rendered HTML.- Greps every
Update <Vendor> <Product>.../Install patch for ...remediation string. - Buckets by leading-vendor token (Mozilla, Microsoft Windows, Microsoft Office family, Notepad++, etc.).
- For each bucket, looks up whether a manifest entry covers it via
matchRemediation(lib/eyas-tools/rtr/library.ts, called throughscripts/rtr-library-cli.mjs match) on a representative string. - Prints two ranked tables: covered (with the script id that handles it, host-count high-to-low) and uncovered (host-count high-to-low, candidate for next-batch curation).
Reading the output
- High-count uncovered → strong candidates for the next
/add-patch-scriptcycle. - Covered but suspicious → e.g. an
app-firefoxrow showing thousands of hits across many version targets is fine; anapp-7ziprow with zero hits means the curated script may have stopped routing (vendor pattern drift) and is worth a manual check via/verify-patch. - Linux kernel CVE descriptions (
Update the last_gc value...,Remove the xhci_initialize_ring_info()...) are noise from grepping CVE description text — the script filters obvious cases but expect a few in the uncovered list.
Note for one-off audits
If the user wants the audit but the dev server isn't reachable (remote work, broken local stack), say so explicitly. There is no offline fallback — Spotlight data is tenant-specific and not in the repo.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

