Agent Gemini CLI
Specialist agent for configuring a new CI pipeline for a supported engine — GitHub Actions or GitLab CI/CD. Resolves the engine target, produces a commit-ready pipeline (hand-authored for GitHub Actions, derived from the platform-neutral capability reference for GitLab), and validates its own output before delivery.
Agent Gemini CLI
Specialist agent for diagnosing and fixing failing GitHub Actions pipelines. Systematically produces a symptom → hypothesis → evidence → fix chain.
Agent Gemini CLI
Specialist agent for operating the project's CI drift-check harness and reconciling the divergences it reports. Runs the harness, interprets its fail-closed output, classifies each divergence by kind, and reconciles it — auto-applying only the deterministic regenerate-and-re-verify case, and diagnosing-and-proposing…
Agent Gemini CLI
Content production specialist. Interviews the product brief, produces a content outline for review, then writes final page copy (hero, features, social proof, CTA). Delivers structured Markdown ready for handoff to frontend-developer or astro-developer. Does NOT make layout or visual decisions.
Agent Gemini CLI
Visual design specialist. Produces color palette tokens, typographic scale, spacing scale, tokens.css, and Tailwind config extensions. Delivers component anatomy specifications and design rationale. Does NOT write application code.
Agent Gemini CLI
Generic implementation agent. Writes, edits, and refactors code with the smallest correct change. Verifies locally before reporting done.
Agent Gemini CLI
Generic documentation agent. Drafts ADRs, READMEs, in-code docstrings, and reference material. Optimizes for documents that age well and stays close to the code where possible.
Agent Gemini CLI
UI implementation specialist. Translates designer token files and component anatomy specs into production-ready CSS and markup. Ensures WCAG 2.1 AA compliance, optimizes asset delivery, and hands off to astro-developer for framework wiring. Does NOT own Astro-specific build concerns.
Agent Gemini CLI
Generic PR and logbook composer agent. Drafts titles, bodies, test plans, logbook entries, and squash-merge commit messages that conform to the project's AGENTS.md conventions.
Agent Gemini CLI
Independent PR reviewer agent. Spawns cold — receives a references-only brief (seat key, PR number, the revision the seat last examined, where its prior verdicts live), never authoring-session context. Activates the pr-reviewer skill to audit the diff, runs linter scripts against changed files, and posts a structured…
Agent Gemini CLI
Runs a smoke or regression pass against a staging or production URL. Diffs results against a stored baseline and surfaces new failures with screenshots and traces.
Agent Gemini CLI
Writes automated test scenarios from a natural-language description of a user journey. Generates Playwright TypeScript or RobotFramework .robot files with page-object model, fixtures, and meaningful assertions.
Agent Gemini CLI
Generic security review agent. Threat modeling, secret hygiene, realistic-threat code review, dependency audit. Findings only — does not implement fixes unless explicitly asked.
Agent Gemini CLI
SEO audit specialist. Audits completeness, Open Graph/Twitter Card tags, structured data (JSON-LD), heading hierarchy, sitemap/robots.txt, internal links, and performance signals with direct SEO impact. Produces a prioritized findings report (critical → high → low). Does not write copy — hands recommendations to…
Agent Gemini CLI
Specification authoring agent. Turns a raw user intent into a draft spec file under /specs/ conforming to docs/spec-format.md, in the interaction mode declared by the parent ticket.
Agent Gemini CLI
Generic test-authoring agent. Writes high-signal regression tests, enumerates priority edge cases, and verifies fixes by failing-then-passing the test against the bug.
Agent Gemini CLI
Detects unintended visual changes between two states of a page by comparing screenshots at multiple viewports. Uses Playwright's toHaveScreenshot or equivalent diff thresholds.
Agent Gemini CLI
Verifies that a page or site matches a given specification — design system, functional requirements, or API contract. Produces a gap report and optionally a Playwright assertion script.
Instructions file GitHub Copilot
Instructions for crewrig/crewrig: This file is the GitHub Copilot CLI entry point for this repository. The agent working rules are defined in AGENTS.md and apply to every agent, including Copilot.
Instructions file CodexOpenCode
Instructions for crewrig/crewrig, covering crewrig — agent working rules, what is crewrig?, session bootstrap, lifecycle and language.
Instructions file
Instructions for crewrig/crewrig, a project described as: A layered AI assistant configuration with shared skill sandbox and built-in harness engineering loop.
Command
Build your personal profile (config/PROFILE.md) through a guided interview. Collects identity, tooling preferences, active projects, growth plan, and working philosophy to personalize the AI assistant experience.
Command
Customize the agent identity file (config/SOUL.md) through guided conversation. Walk the user through each section of the SOUL-E framework (Stance, Origin, Understanding, Lineage, Error Handling) to craft a personalized agent personality.
Skill Claude CodeCodex
Harness feedback-loop curator. Activate on demand to read friction tags from the global harness-friction wing, cluster them, and open one descriptive issue per cluster on the canonical/feedback repos declared in components' provenance blocks. The fix MR lands later (human-authored or via the auto-fix mode tracked in…