feat: LinkedIn suite in tools/, security-audit skill and agent (v4.13.0)

[borghei]

[4.13.0] - 2026-10-07

Added

• New tools/ domain with a 12-skill LinkedIn suite under tools/linkedin/, installable as the tools-skills plugin. The skills cover the whole loop: linkedin-story-interviewer (find what you have to say and keep it in a local story bank), linkedin-content-planner, linkedin-post-writer, linkedin-hook-analyzer, linkedin-humanizer, linkedin-content-repurposer, linkedin-comment-writer, linkedin-reply-manager, linkedin-thread-tracker, linkedin-profile-optimizer, linkedin-employee-advocacy and linkedin-engagement-analytics. 27 stdlib Python tools. The suite is offline by design: nothing posts, schedules or scrapes; you paste drafts yourself and analysis runs on text or exports you provide. Several tools are gates that exit 1 on their deliberately flawed sample input.
• security-audit engineering skill: a six-phase source security audit run by a lead agent with isolated sub-agents — reconnaissance, hunting driven by a coverage ledger, refutation of every candidate by a different agent, schema-checked findings.json, a second fresh read of each record, then reports. Findings end as confirmed, needs validation, or rejected. Six stdlib tools, including validators for the ledger and findings and a safe evidence-promotion tool; 14 references covering 9 core attack classes and 10 domain companions.
• cs-security-auditor agent leads the audit and asks for a breadth tier before every full run: core, focused (core plus web/auth, AI/LLM, supply chain, cloud) or full (all 10 companions). Run it as the session's main agent; it needs to launch sub-agents.

Changed

• Counts across the README, docs and site: 385 skills, 21 domains, 900 Python tools, 77 agents. Engineering skill counts that still said 82 now say 92.

QA (local)

Gate	Result
Tests / lint / build / typecheck	Not configured in this repo (by design); skipped
Script checks	All 33 new scripts compile and answer --help; each runs on its sample in text and JSON and exits 2 on bad input. Security-audit validators reject 28 malformed variants without a traceback; evidence promoter refuses symlinks, hard links, oversize files and traversal
Skill quality scorer	12 LinkedIn skills 100/100; security-audit 86/100
Secret scan	Clean
Security review	0 high, 0 medium. One below-threshold issue fixed: git is now called on the audited repo with core.fsmonitor=false, safe.bareRepository=explicit and --no-optional-locks
Code review (low)	4 findings, all fixed: validator crashes on malformed attempts, unreadable input files, story-bank binding of id-less entries, a dead e.g. pattern. Coverage was partial: the review capped the diff at 160 KB
Site	Rebuilt in the documented order; 385 skill pages, 21 domain pages; no >- descriptions
Migrations	None

Gates in several LinkedIn tools exit 1 on their deliberately flawed sample input; each SKILL.md documents this.

🤖 Generated with Claude Code

https://claude.ai/code/session_019zCKCkv9i8W7d8Vhz2QywP