Agent Claude Code
CLI + TUI UX specialist for ralphctl. Use when designing OR implementing user-facing surface area — command / flag structure, Ink TUI views and prompts, multi-flow session UX, output formatting, error messages, empty-state guidance, help text, theme tokens. Owns src/application/ui/ end-to-end and makes the call on UX…
Agent Claude Code
Documentation custodian for ralphctl. Use when code lands that may have outdated CLAUDE.md or anything under .claude/docs/ (ARCHITECTURE, KERNEL-DESIGN, WORKFLOWS, AI-SETTINGS, SECURITY, PERFORMANCE, REQUIREMENTS, DESIGN-SYSTEM, MANUAL-TEST-PLAYBOOK, HARNESS-PRINCIPLES, RESEARCH-REFERENCES, diagrams/); when a flow…
Agent Claude Code
TypeScript implementer for ralphctl. Use for writing features, fixing bugs, refactoring, or adding tests — anywhere a code change is needed. Respects the four-module Clean Architecture layering (domain → business → integration → application), the chain framework at src/application/chain/, function-first use cases, and…
Agent Claude Code
Implementation planner for ralphctl. Use BEFORE coding begins — when breaking a feature / bug / refactor into scoped, dependency-ordered tasks, identifying affected files, or sanity-checking an approach against the existing architecture. Returns a plan; never writes code.
Agent Claude Code
Prompt template specialist for ralphctl. Use when authoring or editing any .md under src/integration/ai/prompts/ /template.md or src/integration/ai/prompts/partials/, when adding a new placeholder, when adjusting how templates are loaded / substituted (engine/template-loader.ts, engine/substitute.ts…
Agent Claude Code
Code reviewer for ralphctl. Use AFTER implementation — to review a diff / PR / branch for correctness, bugs, architectural-layering violations (domain → business → integration → application), TypeScript nuance (generics, narrowing, Result vs throws), chain composition, and consistency with project conventions.…
Agent Claude Code
Test engineer for ralphctl. Use when writing new vitest tests, shoring up coverage for a module or flow, debugging a flaky / failing test, or designing the test strategy for a new feature. Knows the project's port-based test-double patterns and the flow step-order fence tests.
Hook Claude Code
Runs after a tool call finishes for Edit and Write tool calls, executing format-edited-file.sh. From lukas-grigis/ralphctl.
Settings file Claude Code
Agent settings declaring 1 hook event (PostToolUse) and 7 allowed tools.
Skill Claude CodeCodex
Draft the ralphctl CHANGELOG.md ## [Unreleased] section from the commits since the last release tag — grouping conventional-commit subjects into Keep-a-Changelog sections (Breaking / Added / Changed / Fixed / Removed) and flagging internal churn to omit. Use this whenever you're updating the changelog, preparing…
Skill Claude CodeCodex
Low-level AI CLI spawn mechanics used by ralphctl's provider adapters — headless / interactive spawn, the file-based contract (signals.json + sessionId files), idle-stdout watchdog, exponential rate-limit backoff, and session resume for in-flight recovery. Use when modifying…
Skill Claude CodeCodex
Audit the ralphctl .claude/ setup (the 7 subagents, the in-repo skills, the .claude/docs/ modules) and CLAUDE.md for DRIFT against the real src/ — stale version stamps, renamed/dead paths, removed symbols, unshipped env vars, and fabricated references. Use this whenever someone asks to "audit / refresh / clean up /…
Skill Claude CodeCodex
Check the chain-flow step-traces documented in .claude/docs/ (KERNEL-DESIGN.md examples, the diagrams/, REQUIREMENTS step lists) against the REAL element-name sequence each flow runs, and fix the drift. Use this after changing a flow's element list (adding/removing/reordering a leaf/sequential/guard/loop, renaming an…
Skill Claude CodeCodex
Article-grounded reference for the three Anthropic harness articles. Use when designing or auditing an agent harness, reasoning about long-running-agent state / memory / progress handoff, deciding how much scaffolding a flow needs, pruning components on a model bump, or designing a generator / evaluator split. The…
Skill Claude CodeCodex
Auto-triggers on structural harness decisions: new chain primitive, new flow, remove evaluator, wrap evaluator, redesign harness, scaffolding, load-bearing, model upgrade, harness audit, refactor flow, sub-agent. Also triggers on file mentions of src/application/chain/, src/application/flows/…
Skill Claude CodeCodex
Land an open PR on this repo — wait for CI to finish, then merge with a merge commit using admin bypass. Use whenever the user says "merge this PR", "/merge-pr", "ship the PR", "land the PR", or asks to merge a specific PR number. Defaults to the PR opened from the current branch.
Skill Claude CodeCodex
End-to-end release flow for lukas-grigis/ralphctl. Mirrors the established release pattern: branch → bump → changelog → PR → CI → merge → tag → workflow.
Skill Claude CodeCodex
Run the project's full verification gate — typecheck, lint, tests — in one go. Use after any code change before committing, or when the user asks "is it green?" / "can I ship this?" / "verify" / "run checks".
Instructions file
Instructions for lukas-grigis/ralphctl, covering ralphctl — agent harness for ai coding tasks, build & run, read on demand, architecture invariants and implementation style.
Skill Claude CodeCodex
Cross-phase skill — design the shape of the change (entities, boundaries, seams) before generating code, tasks, or acceptance criteria. Failure mode is "big blob" output that obscures the core change.
Skill Claude CodeCodex
Cross-phase skill — establish a shared understanding of what will and will not be done before producing output. Restate the input back to the user, surface assumptions, name non-goals, and agree before you write. For an input that is still a raw, unshaped idea needing multiple candidate directions before one is…
Skill Claude CodeCodex
Session-discipline skill — settle the plan before touching code, work in small verified increments with a feedback loop that proves each change works, and record corrections so the same mistake is not repeated. Use when executing multi-step coding tasks where batching unverified work would compound risk; bundles…
Skill Claude CodeCodex
Multi-phase code-quality skill — primary frame for the evaluator role in Execute, the architecture axis in Plan, and correctness/readability in Refine. Multi-axis code review with severity vocabulary. Use when you are the evaluator assessing a generator's output, and when reviewing any change before signalling…
Skill Claude CodeCodex
Systematic root-cause debugging. Use when tests fail, builds break, or behaviour does not match expectations. Follow stop-the-line → reproduce → localize → reduce → root-cause → guard-with-regression-test → verify, not guessing; the reproduction and regression steps follow the same red-green discipline as…