Three new specialist agents filling clear gaps in the current roster:
- engineering/engineering-data-engineer.md: ETL/ELT pipelines, Delta Lake,
Apache Spark, dbt, Medallion Architecture (Bronze/Silver/Gold), streaming
with Kafka, data quality with Great Expectations, cloud lakehouse platforms
(Fabric, Databricks, Synapse, Snowflake, BigQuery)
- engineering/engineering-technical-writer.md: Developer documentation,
API reference (OpenAPI), README templates, tutorials, docs-as-code
(Docusaurus/MkDocs), content quality standards, DX metrics
- specialized/specialized-developer-advocate.md: Developer relations,
DX auditing and improvement, technical content creation, community
building, GitHub engagement, conference talks, product feedback loops
Each agent follows the standard template with YAML frontmatter, identity,
core mission, critical rules, technical deliverables with working code,
workflow process, communication style, and success metrics.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
The installer UI preview in README.md contained Unicode box-drawing
characters (╔╗╚╝║═) and bullets (●○✓) from an earlier draft that
didn't match the actual script output. Replaced with the current
plain-ASCII UI output that the script actually renders.
Also replaced em-dashes (—) with -- outside of code blocks for
consistency with the rest of the documentation.
The 250+ generated files in integrations/ were adding repo bloat and would
cause merge conflicts when agents change. Run scripts/convert.sh to
regenerate them locally.
What's committed:
- scripts/convert.sh and scripts/install.sh (the actual value)
- integrations/*/README.md (docs for each tool)
- integrations/README.md (overview)
- .gitignore updated to exclude all generated output dirs
What's excluded (add to .gitignore):
- integrations/antigravity/agency-*/
- integrations/gemini-cli/skills/ + gemini-extension.json
- integrations/opencode/agent/
- integrations/cursor/rules/
- integrations/aider/CONVENTIONS.md
- integrations/windsurf/.windsurfrules
Responds to issue #47 — blesses jnMetaCode's zh-CN localization repo
and adds a Community Translations & Localizations section to the README.
Closes#47
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Excellent contribution! The Accessibility Auditor fills an important gap — great depth with the audit report template, screen reader testing protocol, and keyboard navigation audit. The axe-core examples are a nice practical touch. We'll handle the README update on our end. Thanks @benjifriedman!
Great agent! The Security Engineer is thorough — love the STRIDE threat model template and the concrete CI security pipeline examples. We'll handle the README update on our end. Thanks @jnMetaCode!
These workflow examples are a great addition — practical, well-structured, and aligned with the roadmap. The startup MVP and landing page sprint patterns will be really useful references. Thanks @jnMetaCode!
Excellent addition! The lint workflow looks solid — validates required frontmatter as errors, recommended sections as warnings, and only lints changed files to avoid noise. Thanks @jnMetaCode!
Add Xiaohongshu Specialist, WeChat Official Account Manager, and Zhihu
Strategist to the Marketing Division roster. Update agent count to 59.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Demonstrates 8 agents (Product Trend Researcher, Backend Architect,
Brand Guardian, Growth Hacker, Support Responder, UX Researcher,
Project Shepherd, XR Interface Architect) deployed in parallel to
evaluate a software opportunity and produce a unified product plan.
Adds examples/ directory with README and the Nexus Spatial discovery
document. Links from the project README under a new Scenario 4 and
checks off the multi-agent workflow examples roadmap item.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Fix ((errors++)) causing early exit under set -e
- Fix command injection in workflow by using env variable
- Add strategy/ directory to lint paths and AGENT_DIRS
Adds two step-by-step workflow examples showing how to coordinate
multiple agents for real projects:
- Startup MVP: 7 agents across a 4-week build cycle
- Landing Page Sprint: 4 agents shipping a page in one day
Each example includes exact prompts to copy, handoff patterns, and
tips for passing context between agents. Checks off the
"Multi-agent workflow examples" item from the roadmap.
- Add missing color field to 4 agents: content-creator, growth-hacker,
social-media-strategist, data-analytics-reporter
- Replace non-standard tools field with the required color field
- Fix CRLF line endings in agentic-identity-trust.md
Adds a specialized Security Engineer agent covering threat modeling,
vulnerability assessment, secure code review, and security architecture.
This fills a gap in the Engineering Division where no dedicated security
specialist existed.
Adds a lint script and GitHub Actions workflow that checks agent
markdown files for required YAML frontmatter fields (name, description,
color) and recommends standard sections. Only changed files in PRs
are validated to avoid blocking on pre-existing issues.
Remove unrestricted Bash tool access from 7 agents that only need
analytical/advisory capabilities, rewrite the Social Media Strategist
agent (was a duplicate of Twitter Engager) to cover multi-platform
strategy as intended, fix incorrect descriptions, harden webhook
example to use env vars, and clarify ambiguous AMA language.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
