v1.0.13

Adds a persistent Voice & Brand Profile that every writing skill reads before drafting, so you stop re-explaining who you are on each request.

• references/voice-profile.md — your voice fingerprint, niche/ICP, hard rules, links/CTA, and signature examples in one file. Ships as a template; skills only load it once you mark it filled: yes.
• --mode profile — builds/updates the profile from 3-6 of your real posts pasted in (portable, no token) or, if a read token is set, from pulled activity. Learns your actual fingerprint, never invents one.
• The bundle's writing skills load the filled profile and match your voice, rules, and CTA automatically.

Universal and cross-platform: same behavior on Claude Code, Codex, and OpenClaw (no Anthropic-only primitives).

v1.0.12

Doc consistency: the remaining '10 X formulas' references inside x-post-writer, x-hook-extractor, and x-repurposer now read 11 (X11 was added in v1.0.10). No behavior change.

v1.0.11

Doc fix: README now says 11 X hook formulas (X11 Third-Person Case-Study Flex was added in v1.0.10). No behavior change.

v1.0.10

Corpus-validated the X hook formulas against ~450 top English tweets (Apify, normalized by follower count, top 69 hand-checked).

• Added X11 Third-Person Case-Study Flex — narrating someone else's result ("this person made $X doing Y") was the single most common winning shape and had no formula.
• Reversed the "How I beats How to" rule — first-person teardowns were near-absent among real over-performers; third-person case studies dominated.
• Strengthened X2 with the corroborated big-brand-spend + same-day-consequence variant.
• Added ironic day-in-life list + meme-juxtaposition as X6 variants; flagged like-for-DM engagement bait as not-to-copy.
• Propagated X11 into x-post-writer + x-hook-extractor classification.

v1.0.9

New skill: Audience Insights (9th) + an Apify read layer.

Reads your X audience and niche from real data (verified live against the actor): pull a handle's recent tweets with likes/replies/views, see the pattern behind the winners, read the repliers on a tweet (X gates likers, so repliers are the signal), and scan a niche for top tweets. Normalizes engagement by follower count so a small account's breakout is not buried. No login; ~$0.15/1k tweets, or paste your own data. Patch release.

v1.0.8

New skill: Profile Optimizer (8th in the bundle).

Audits and rewrites the seven parts of an X profile that decide a follow in the two-second profile-card glance: bio (160 chars), searchable display name, @handle, header, pinned tweet, link, and photo. Delivers a scorecard, priority fixes, before/after rewrites, and a pinned-tweet pick matched to your goal.

Turn a default profile into one that converts visitors into followers. Patch release.

v1.0.7

Patch: trim the new Repurposer skill's description under the 510-char cap. No behavior change.

v1.0.6

New skill: Repurposer (7th in the bundle).

Turn content you already made on another platform into a native X post or thread. It transforms, it does not copy-paste: takes a LinkedIn post, blog paragraph, YouTube script, or newsletter, extracts the spine, re-hooks for X's no-fold first line, refits to 280 chars or a thread, strips off-platform artifacts (link in bio, hashtag walls, smash subscribe), runs the humanizer, and waits for approval.

Write once, publish everywhere natively. Patch release.

v1.0.5

Cross-runtime install docs: the README now documents OpenClaw and Hermes Agent install paths (Hermes reads SKILL.md per the agentskills.io standard and imports OpenClaw skills via hermes claw migrate), alongside the existing Codex, Claude Code, Claude Desktop, claude.ai, and skills-CLI options. Docs-only patch, no skill changes.

v1.0.4

Hardening patch (parity with linkedin-skills v1.0.10/v1.0.8 line).

• Security pack: SECURITY.md, HOL plugin-scanner CI workflow, dependabot config, requirements lockfile, .codexignore, and a branded plugin icon wired into the Codex manifest.
• Audit robustness verified: the pre-publish audit was pressure-tested (authority + deadline scenario, platform-adapted AI-tell drafts) and correctly FAILED the bad draft with full blocker lists. No rule gaps found.
• Writers: goal-first selection added where missing (YouTube community post writer, TikTok caption writer in their respective bundles).

No breaking changes. Patch release.