* fix(youtube-summarizer): close unclosed code fence in Step 5 output section
The ```markdown output structure template in Step 5 was never closed.
The closing ``` was missing after the [Final synthesis and takeaways]
placeholder, causing all subsequent example sections (Example 2,
Example 4, Executive Summary) to render as verbatim text inside the
code block rather than as structured markdown sections.
Co-Authored-By: Claude Code <noreply@anthropic.com>
* fix(youtube-summarizer): move code-fence fix to source skill path
---------
Co-authored-by: claude[bot] <claude[bot]@users.noreply.github.com>
Co-authored-by: Claude Code <noreply@anthropic.com>
Co-authored-by: sickn33 <sickn33@users.noreply.github.com>
* fix(prompt-engineer): add missing Step 2 to workflow sequence
The workflow jumped from Step 1 (Analyze Intent) directly to Step 3
(Select Framework), skipping Step 2 entirely. All cross-references in
the ALWAYS rules and examples already named Step 2 as "Ask Clarifying
Questions (Conditional)" — this commit adds that step with its trigger
conditions, question limits, and a clarifying exchange example.
Co-Authored-By: Claude Code <noreply@anthropic.com>
* fix(prompt-engineer): move Step 2 fix to source skill path
---------
Co-authored-by: claude[bot] <claude[bot]@users.noreply.github.com>
Co-authored-by: Claude Code <noreply@anthropic.com>
Co-authored-by: sickn33 <sickn33@users.noreply.github.com>
* feat(skill): Added a skill for agents understand how then can work with mise-en-place
* Update related skills in SKILL.md
* fix(skill): Make mise configurator deterministic
* chore(package): add a npm run command, to sync skills of automatic maner on README.md
* feat(tool): add a new tool script to count and replace number of script on README.md
* refactor(method): adjusted method to grant a return
---------
Co-authored-by: sickn33 <sickn33@users.noreply.github.com>
* feat: upgrade IT skills to Digital Governance Suite with COBIT, TOGAF, NIST and SRE
* fix: update persona count in it-manager-pro to match expert list
---------
Co-authored-by: edudeftones-cloud <edudeftones-cloud@users.noreply.github.com>
* feat: add not-human-search-mcp and ai-dev-jobs-mcp skills
Two remote MCP server skills for AI agent runtime discovery:
- not-human-search-mcp: Search/score 1,750+ AI-ready websites, verify MCP
endpoints via JSON-RPC probe. Endpoint: https://nothumansearch.ai/mcp
- ai-dev-jobs-mcp: Search AI/ML jobs from 289+ companies, get job details,
list companies, market stats. Endpoint: https://aidevboard.com/mcp
Both are free, require no authentication, and use streamable HTTP transport.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(skills): Align MCP tool docs with live endpoints
---------
Co-authored-by: Owl <owl@arflow.io>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: sickn33 <sickn33@users.noreply.github.com>
* feat(skill): Added a skill for agents understand how then can work with mise-en-place
* Update related skills in SKILL.md
* fix(skill): Make mise configurator deterministic
---------
Co-authored-by: sickn33 <sickn33@users.noreply.github.com>
Record the 10.1.0 release summary covering PRs #514, #516, #517, the direct fix for issue #518, and the post-10.0.0 limitations audit hardening.
Refs #514
Refs #516
Refs #517
Refs #518
