update-threat-db
Update the guide's threat intelligence through AgentSec Triage, research new coding-agent and MCP security advisories, and synchronize the security feed. Use for threat database refreshes or mise a jour de la base de menaces, not repository security scans or Claude Code release tracking.
- License 许可
- MIT
- 最近更新
- 2026-10-18
- Stars
- ★ 0
- Forks
- 2
01 前言 Front Matter
- name
- update-threat-db
- description
- Update the guide's threat intelligence through AgentSec Triage, research new coding-agent and MCP security advisories, and synchronize the security feed. Use for threat database refreshes or mise a jour de la base de menaces, not repository security scans or Claude Code release tracking.
02 资源文件 1
- 参考文档evals/scenarios.json1.5 KB
03 SKILL.md 正文
Update threat intelligence through AgentSec
Scope: this guide repository and its AgentSec and landing consumers. This skill
does not change global agent configuration. Its canonical source is
.claude/skills/update-threat-db/; regenerate the Codex projection with
python3 scripts/sync-threat-skill.py --write from the guide root.
Resolve the workflow
Read .claude/commands/update-threat-db.md from the guide root in full. It owns
the delegation contract and consumer checks. Resolve AGENTSEC_REPO if set,
otherwise the sibling ../agentsec-triage. Verify the required files before
continuing. Do not silently substitute the guide's compatibility database.
Read the resolved AgentSec AGENTS.md and
.claude/commands/update-threat-db.md in full, then execute that workflow.
Preserve unrelated edits in all checkouts. Use the latest recorded update date
through today's date as the research window.
Evidence and changes
Review primary sources before accepting new records. Keep the evidence sources, dated events, exact detector inputs and public feed as separate artifacts. Record affected ranges, fixed versions and uncertainty only when the source supports them. Deduplicate CVE/GHSA aliases. Keep corrections as dated events.
Add the required regression test before each accepted data change. Distinguish
detected, partial, not_detected and not_applicable; a documented CVE does
not prove executable detection. Do not run malicious samples.
Completion checks
- Pass the AgentSec validation gates and regenerate artifacts twice without drift.
- Synchronize the approved public feed into the guide and landing; verify byte equality and run each consumer's integration checks.
- Refresh the guide's compatibility database and reader guidance only after the canonical records pass. Update each affected repository's changelog.
- Run
python3 scripts/sync-threat-skill.py --checkafter editing this skill or its routing scenarios. Validate both host routes with the installed router. - Report the research window, accepted and deferred records, detector limits, file changes, validation results and synchronization state.
Commit, push, tagging and publication are separate states. Follow AgentSec's authorization and data-license boundaries; a local update does not authorize a public release. State any unfinished gate explicitly.
04 同仓库更多 Skills
voice-refine
Transform verbose voice input into structured, token-efficient Claude prompts. Use when cleaning up voice memos, dictation output, or speech-to-text transcriptions that contain filler words, repetitions, and unstructured thoughts.
validate-changes
Evaluate staged changes using LLM-as-a-Judge before committing
update-threat-db
Delegate threat-intelligence research and updates to AgentSec, then validate the guide and landing mirrors.
token-audit
Audit Claude Code configuration to measure fixed-context token overhead and produce a prioritized action plan. Use when hitting rate limits, experiencing early context compression, or after adding significant config files.
talk-stage6-revision
Produces revision sheets with quick navigation by act, a master concept-to-URL table, Q&A cheat-sheet with 6-10 anticipated questions, glossary, and external resources list. Use when preparing for a talk with Q&A, creating shareable reference material for attendees, or building a safety-net glossary for live delivery.
talk-stage5-script
Produces a complete 5-act pitch with speaker notes, a slide-by-slide specification, and a ready-to-paste Kimi prompt for AI slide generation. Requires validated angle and title from Stage 4. Use when you have a confirmed talk angle and need the full script, slide spec, and AI-generated presentation prompt.