advisory-version-check
Skill0 starsChanged 5 months agoMITView on GitHub
Compare one local stack inventory against one advisory's affected version ranges. Use when an advisory names a component and the question is whether this site's recorded version falls inside the stated range. Read-only, and never a safety verdict.
In this repository83 other files it gives its agents.
- AGENTS.mdAGENTS.md
- docs/AGENTS.mdAGENTS.md
- shep/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- shep/CLAUDE.mdCLAUDE.md
- aeo-llm-content-plannerSkill
- agency-command-centerSkill
- agency-morning-briefSkill
- agency-scorecardSkill
- agent-runbookSkill
- agent-usageSkill
- autonomy-budgetSkill
- backlink-pr-angle-generatorSkill
- brand-kit-from-urlSkill
- browser-automationSkill
- building-autonomous-agentsSkill
- capability-mapSkill
- case-study-engineSkill
- clickup-apiSkill
- client-dashboardSkill
- client-health-scoreSkill
- client-onboarding-osSkill
- cli-llm-routingSkill
- codebase-mapSkill
- command-from-your-phoneSkill
- competitor-content-gap-analysisSkill
- connection-monitorSkill
- content-brief-generatorSkill
- determinism-patternSkill
- dispatch-and-briefSkill
- division-operating-modelSkill
- dreamingSkill
- durable-stateSkill
- entity-topical-authority-mapperSkill
- first-real-jobSkill
- founding-circle-ai-securitySkill
- gbp-competitor-auditSkill
- goal-modeSkill
- host-your-agentSkill
- html-reportsSkill
- instinct-bridgeSkill
- evolveSkill
- learn-evalSkill
- learnSkill
- promoteSkill
- skill-createSkill
- internal-linking-auditorSkill
- keyword-deep-diveSkill
- ladder-auditSkill
- lanes-momentum-frameworkSkill
- launch-websiteSkill
- llm-citation-auditSkill
- migration-seo-paritySkill
- migration-visual-checkSkill
- model-gear-routerSkill
- multi-account-gatewaySkill
- multi-model-councilSkill
- night-shiftSkill
- novaSkill
- per-client-memory-routerSkill
- The first 60 shown.
The file
Files
advisory-version-check
The problem this solves: an advisory says "PHP before 8.2.10" and the site runs 8.2.4. Someone still has to compare the two. This skill does that comparison and nothing else.
It sits after php-advisory-watch and stack-inventory. The first says an advisory exists. The second lists what the site runs. This one compares a version against the range the operator writes down.
Say this to your agent
"Compare this inventory against this advisory. Use only the two JSON files I give you. Mark each named component in-range, out-of-range, not-present, or UNKNOWN. Do not contact the site, do not fetch the advisory, and do not say the site is safe."
What the operator supplies
Two local JSON files.
The inventory is the file stack-inventory writes. The advisory file is written by the operator from the advisory they already read:
{
"id": "GHSA-example",
"affected": [
{"name": "php", "range": ">=8.2.0,<8.2.10"}
]
}A range is comma-separated clauses: >=X, >X, <=X, <X, or ==X. A bare number means an exact match.
What it writes
One JSON report on stdout:
in-range: the recorded version falls inside the stated rangeout-of-range: it falls outside that rangenot-present: the inventory has no component by that nameUNKNOWN: the version or the range could not be read
Exit 0 when every row was comparable. Exit 2 when any row is UNKNOWN or either file cannot be read. out-of-range is not "safe." It only means the recorded version is outside the range the operator typed.
Run it
python3 skills/advisory-version-check/templates/version_check.py \
--inventory stack-inventory.json \
--advisory advisory.jsonWhat it will not do
- It does not fetch an advisory, scan a site, or log in.
- It does not upgrade anything or propose a patch command.
- It does not treat a missing component or an unreadable version as safe.
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.

