By tomharris
Automates senior software engineer tasks: PR review, Slack Q&A, ticket implementation, design doc review, and standup generation. Uses an approval queue so you stay in control.
Register the current project with engineer-agent
Manually add a Jira ticket or GitHub issue to the implementation queue
Scan a project for bugs and security issues; verified findings enter the review queue
Detect gaps between spec, design doc, and tickets in the pipeline
Create a design doc from a refined PM feature spec in Slite
Draft an answer to a Slack question about code, implementation, or project status. Use this skill when processing a slack-question queue item or when asked to draft a Slack response.
Scan a project (or subdirectory) for bugs and security issues. Uses a Sonnet subagent to surface candidate findings, then an Opus subagent to verify each candidate. Verified findings become code-audit-finding queue items and trigger a ntfy push if configured. Use this skill when processing an /engineer-agent audit-code invocation.
Compare pipeline artifacts across a boundary (spec↔design or design↔tickets) and produce a checklist of gaps. Use this skill when processing a gap-audit queue item.
Generate a structured engineering design doc from a refined PM feature spec. Use this skill when processing a design-doc queue item.
Break a design doc into phased implementation tickets with dependencies. Use this skill when processing a ticket-plan queue item. Supports both Jira and GitHub Issues as the target tracker.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A Claude Code plugin that automates senior software engineer tasks with an approval-gated workflow. The agent polls GitHub, Slack, ticket trackers (Jira or GitHub Issues), and Slite for work, drafts responses, and queues everything for human review before posting externally.
Everything goes through an approval queue — nothing is posted until you say so. One unified queue across all your projects. With ntfy configured, that approval can happen remotely from your phone, and safe actions (draft-PR creation) can be allowed to run hands-free.
gh) — for PR reviews, code operations, and GitHub API access. Install from cli.github.com and run gh auth loginspy) — for channel polling and message posting. Spy reuses your local Slack desktop session, so there's no OAuth or app install. Build it from github.com/tomharris/spy (go install ./cmd/spy), sign in to the Slack desktop app, then run spy auth to confirm. Set the workspace via agent.slack.workspace in config (required when multiple workspaces are signed in)mcp__atlassian__*). Either Jira or GitHub Issues per project — GitHub Issues uses the gh CLI (already a prerequisite)mcp__slite__*)curl against ntfy.sh or your self-hosted server); install the ntfy mobile app to receive alerts and tap Approve/Reject. The approval listener also needs jq on the host running it.# For development (loads plugin for this session only)
claude --plugin-dir /path/to/engineer-agent
# For permanent installation
# Inside a Claude Code session, run:
# /plugin marketplace add /path/to/engineer-agent
# /plugin install engineer-agent
Then initialize from any project directory:
/engineer-agent setup
This creates the user-level config, queue directories, and installs automated polling in one step. It also registers the current project. After setup, edit ~/.claude/engineer-agent/engineer.yaml to configure your sources.
To register additional projects:
cd /path/to/another-project
/engineer-agent add-project
Edit ~/.claude/engineer-agent/engineer.yaml:
npx claudepluginhub tomharris/engineer-agent --plugin engineer-agentComprehensive UI/UX design plugin for mobile (iOS, Android, React Native) and web applications with design systems, accessibility, and modern patterns
Standalone image generation plugin using Nano Banana MCP server. Generates and edits images, icons, diagrams, patterns, and visual assets via Gemini image models. No Gemini CLI dependency required.
Multi-model consensus engine integrating OpenAI Codex CLI, Gemini CLI, and Claude CLI for collaborative code review and problem-solving.
Write feature specs, plan roadmaps, and synthesize user research faster. Keep stakeholders updated and stay ahead of the competitive landscape.