ingest-website — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited ingest-website (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Full lifecycle for a single website task: fetch via Jina Reader → summarise → write report → append suggestion → mark done.
Prerequisites: Network access tor.jina.ai. For Google Tasks API calls, refer to../gws-tasks/SKILL.md.
Prepend https://r.jina.ai/ to the target URL and fetch using read_url_content:
https://r.jina.ai/<TARGET_URL>If Jina Reader fails (empty response, HTTP error, or clearly broken content):
📄 Fall back to the content-cleaner skill to perform a direct HTTP fetch with AI text extraction.If both fail, log the error and skip — do NOT mark the task as completed:
⚠️ Skipping '<title>': fetch failed (Jina + content-cleaner both failed).📄 Read ../content-summary/references/summarise.md📄 Read ../content-summary/references/filename_rules.mdDirectory: reports/Website_YYYY_MM_DD/ Filename: [domain]_[slugified_title].md (hostname without www., then slugified page title)
📄 Read ../content-summary/references/output_template.mdWebsite-specific fields under ## 🔖 來源 Metadata:
martinfowler.com)Omit `## 📄 原始內容` entirely — do not include the raw Jina Markdown in the report.
Confirm the file is written before proceeding.
📄 Read ../content-summary/references/ai_analysis.md📄 Follow ../content-summary/references/suggestion_log.md{SourceType} = Website
gws tasks tasks patch \
--params '{"tasklist": "<DELEGATE_LIST_ID>", "task": "<TASK_ID>"}' \
--json '{"status": "completed"}'Log: "✅ Task '<title>' marked as completed. Report saved to reports/Website_YYYY_MM_DD/<filename>.md"
Jina returns empty or garbled content: Dynamic React/SPA sites may not render well. Fall back to content-cleaner and proceed with whatever text is extracted.
`gws tasks tasks patch` fails: Double-check tasklist and task are IDs (not titles).
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.