Claude Code plugin validator rejects bare ./ prefixes as escaping
the plugin directory. Strips ./ from all skill paths and source field
so the plugin loads correctly.
Closes#243
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The `hooks send` command accepted any user-supplied URL without validation.
This adds an https:// scheme check and a comment clarifying the intended use
(Zapier webhook catch hooks), matching Zapier's own webhook URL format.
Co-Authored-By: Claude Code <noreply@anthropic.com>
The unquoted description contained "Trigger phrases:" which YAML parsed
as a nested mapping, causing the skills CLI to silently drop the skill
during discovery. Wrap the value in double quotes and escape inner
quotes.
The Zapier SDK (@zapier/zapier-sdk) gives AI agents direct TypeScript
access to 8,000+ app integrations without building OAuth flows. Updates
the integration guide with SDK quick start, CLI commands, methods, and
marketing automation examples.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Move full content refresh content to references/content-refresh.md
to stay under 500 lines when merged with expert panel PR
- Bump version to 1.3.0 (expert panel PR takes 1.2.0)
- Fix scope creep: remove structured data from step 6, keep it
focused on scannable formatting
- Add content-refresh.md to References section
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Adds guidance for refreshing existing content — when to refresh vs.
rewrite, a 6-step refresh checklist (freshness, accuracy, voice, SEO,
proof, structure), and recommended cadence by content type. Bumps to v1.2.0.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Gong: fix endpoints to POST (not GET), remove non-existent /v2/deals,
use tenant-specific base URL, correct recording limitation, add source
links. RB2B: clarify API Partner Program vs native integrations, add
pricing volatility note and source links. Registry: fix '10k calls'
to '10k API calls'.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add bridge sentence connecting to Seven Sweeps workflow
- Fix scoring criteria conflict: all personas 7+ with 8+ average
- Aligns step 5 target with the scoring rubric
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Soften unsourced "85%" stat to "most social video"
- Add "or audiograms" for audio-only podcast repurposing
- Broaden content atoms intro to apply across all formats
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Soften win-rate language to context-aware guidance. Clarify weekly
cadence to monitor guardrails while avoiding premature winner calls.
Expand playbook template with sample size, CI, guardrails, segment
deltas, and implementation status.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Multi-persona review technique for high-stakes copy. Assemble 3-5 expert
personas, score 1-10, iterate until 8+. Includes recommended panels for
landing pages, email sequences, and sales pages. Bumps to v1.2.0.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Corey Haines