heart-of-gold-toolkit

Survey a repo and ground externally on the relevant parts of the internet. Local manifest scan + CLAUDE.md/AGENTS.md + recent git activity, then external grounding via WebSearch, WebFetch, and gh — synthesized into a terse, dated, sourced briefing cached at docs/ground/<fingerprint>.md with a 7-day freshness window. Composable: invoked standalone or embedded as a pre-flight phase in other skills. Triggers: ground, ground yourself, ground myself, survey and ground, pre-flight grounding, research the stack, what's new in.

Collaborative discovery before planning. Explore the problem space, evaluate approaches, surface past work, and produce a structured brainstorm document. Triggers: brainstorm, explore, discovery, ideate, think through, what should we build, explore approaches.

Strategic planning with auto-calibrated detail, decision rationale, and dependency ordering. Use when starting a new feature, bug fix, refactor, or any non-trivial work. Produces a plan document with tasks, reasoning, and acceptance criteria. Triggers: plan, planning, create plan, implementation plan, feature plan, work plan.

Execute a plan from start to ship — read tasks, implement in dependency order, test continuously, commit incrementally, run quality checks, and push. The plan's checkboxes are the tracker. Triggers: work, execute plan, implement, start work, build, ship, finalize, release, push, ready to ship, done building.

Capture a footgun the agent just hit (or the user just spotted) as a proposed trigger via `quellis teach`. Guided flow: gather the finding, run the CLI, surface the proposed trigger, leave activation to the user. Triggers: teach quellis, capture this footgun, add a trigger, remember this rule.

Review and manage evidence contracts in this repo. Subcommands: `show` (default, print the active contract), `list` (all contracts + which is active), `complete` (mark active contract completed), `abandon` (mark active contract abandoned). Triggers: contract, show contract, list contracts, mark contract complete, what's active.

Narrowly-scoped problem framing for a single coding task. Use before `/quellis-architect:plan` when the task shape is fuzzy — what's the smallest version that ships value, what are the load-bearing constraints, what evidence will count as done. Triggers: brainstorm, frame, scope, what's the smallest version, how should I cut this.

Produce the evidence contract for a single coding task. Decomposes the task into work units and names the evidence each claim will require at Stop time. Output is `.quellis/contracts/<task-id>.toml` (contract.v1). Triggers: plan this task, write the contract, what's the evidence, plan with quellis.

Walk the active evidence contract claim-by-claim and gather the evidence each claim requires. Run before Stop. Produces a verify report; does not bypass the Stop hook. Triggers: verify, check the contract, run the verify pass, am I actually done.

Install harness-engineering doctrine into a repository — short root AGENTS.md, docs/ taxonomy (plans, ADRs, solutions), verification rules, and project-scoped Claude Code plugins. Works on empty repos (greenfield scaffold), existing repos (survey + upgrade), and ports of existing systems. Distinct from `marvin:scaffold` (which installs deps and configs); distinct from `cc-lab:cc-lab-diagnose` (which observes but doesn't change anything). Triggers: harness up, harness this repo, set up AGENTS.md, agent doctrine, make this repo agent-ready, init harness, scaffold agent context.

Use when the user asks to run Codex CLI (codex exec, codex resume) or references OpenAI Codex for code analysis, refactoring, or automated editing

AI image generation and editing. Text-to-image, style transfer, and logo generation. Powered by GPT Image 2 (via Codex CLI, no API key), Gemini, and FLUX. Triggers: generate image, create image, make image, draw, illustrate, logo, visual, picture.

Strategy-forward product cartography. Articulates the goal, grounds the analysis in domain research, commits to a strategic thesis, then maps three mutually-exclusive futures that execute it — each with a user's love-letter, required capability, what it excludes, and a next move you can hand to your agent. Auto-invokes /babel-fish:visualize for the visual map. Use when exploring "where could this go," forcing a strategic commitment, or stress-testing a product idea for lovability. Triggers: improbable futures, product futures, where could this go, map my project, lovable product, cartographer, product atlas, product strategy.

Turn a set of screenshots or images into a LinkedIn document-post PDF (carousel) with per-page background matching so padding disappears into the source. Triggers: linkedin carousel, carousel pdf, document post, slide pdf, screenshots to pdf, linkedin pdf, carousel from images.

Create shareable HTML visual artifacts from markdown, plans, architecture docs, brainstorms, and other structured content. Prefer browser-viewable HTML first when it will materially improve clarity or sharing; otherwise fall back to terminal rendering. Triggers: visualize, mindmap, mind map, show me the structure, draw a map, make this clear, make this visual.

Control the local share server lifecycle after setup. Use when the user wants to check status, start or stop the local share server, restart it, or enable or disable private Tailscale viewer exposure.

Turn brainstorm decisions into user stories, architecture doc, and ADRs. Use after brainstorming to define WHAT to build and HOW it fits together. Standalone or pipeline-aware via env vars. Triggers: architect, stories, architecture, design, ADR, define stories, write stories.

Publish a local HTML file or static site directory to a configured local share server and return a browser-viewable URL. Use when the user wants to open coding-agent output in a browser on the same machine or another device over a private tailnet.

Set up a local share server for coding-agent HTML/static artifacts, or connect portable share skills to an existing compatible server. Use when the user wants local browser-share infrastructure, Tailscale-backed viewing, or reusable artifact publishing across coding agents.

Two-layer copy editor. Layer 1 is a deterministic typography audit (regex-level, auto-closeable). Layer 2 is LLM judgment — reject-list hits, nominal-style detection, clarity/ambiguity pass for participant-facing content, voice/register check, and spoken- readability read. Loads repo-local .copy-editor.yaml to compose the baked-in language profile with the repo's style guide, reject list, examples, and voice doctrine. Czech is a full profile; English is a stub. **Activate automatically when the conversation edits or creates any file matching a repo's .copy-editor.yaml include scope, when the user asks about Czech or participant-facing copy, when reviewing content PRs, or when a work slice would otherwise close without a Layer 2 pass on edited visible-surface content.** Do not wait to be invoked by name. Explicit trigger phrases: copy edit, editorial pass, czech review, czech copy review, audit copy, check typography, review prose, tighten copy.

ICF coaching conversation — powerful questions, anti-sycophancy, discovery

Quick accountability check — commitment, progress, milestones, obstacles

Guided goal-setting using SMART+V framework, values alignment, and obstacle pre-mortem

Guided reflection using FLOW, REVIEW, SCARF, and emotion handling frameworks

Multi-framework content review with convergent synthesis. Each lens applies a named analytical framework — grounded in published source material, not persona impersonation — to the target content. Parallel dispatch, web research or user-provided references, convergent synthesis, prioritized action list. Triggers: expert panel, expert review, expert audit, panel review, multi-lens review, framework review, content audit, workshop audit, strategy review, get expert feedback, advisory review.

Adversarial review — find weaknesses and expose them with failing tests. Checks architecture conformance, stub detection, security, and story completeness. Never modifies implementation. Standalone or pipeline-aware via env vars. Triggers: redteam, adversarial review, attack, probe, find weaknesses.

Use when the user asks to run Claude Code CLI (`claude`, `claude resume`) for review, analysis, implementation, refactoring, debugging, or follow-up specifically through Claude Code.

Focused review of code, documents, or architecture — one deep pass with evidence-based findings and clear verdict. Auto-detects what you're reviewing: branch diff, PR, file path, plan, brainstorm, or spec. One reviewer that reads carefully beats nine that skim. Triggers: review, code review, review PR, review diff, review plan, review brainstorm, review spec, review document, evaluate, check.

Capture knowledge — solutions, context docs, learnings, and principles. Use after fixing non-trivial bugs, creating context for AI, or discovering patterns worth preserving. Triggers: compound, document solution, capture fix, save solution, knowledge compound, document this, save this fix, context, create context, update context, build context, learn, save learning, remember this.

Generate audio content — text-to-speech, podcasts, voice cloning, sound effects, speech-to-speech, dubbing, and audio isolation. Currently powered by ElevenLabs. Works with both the Python SDK and the ElevenLabs CLI. Includes ready-to-run generator scripts that Claude writes to a temp file and executes directly. Triggers: audio, elevenlabs, text-to-speech, TTS, podcast, voice, voiceover, narration, voice clone, sound effects, dubbing, speech-to-speech, audio isolation.

Capture what's on your mind — morning intentions or evening reflections — and save a structured daily thought record

Run the daily content pipeline to fetch signals, analyze relevance, draft output, edit for voice fidelity, and deliver the brief

Configure the Guide content pipeline by creating or updating content/config.yaml, sources, themes, and voice profile settings

Guided 7-phase blog post writing. Use when writing a long-form blog post from scratch or from a pipeline-generated outline. Triggers: write post, blog post, write article.

Fast opinionated quality pass — simplicity, test integrity, and correctness. Not a deep review. One pass, one voice. For depth, use /review (deep-thought). Triggers: quick review, quality check, sanity check, check my code, is this good.

Prepare a project for development — create directories, configs, install dependencies. Never creates source or test files. Standalone or pipeline-aware via $ARCH_PATH. Triggers: scaffold, setup project, prepare project, init project, create structure.

Write failing tests from user stories and architecture docs. Behavioral tests verify WHAT (acceptance criteria), conformance tests verify HOW (architecture compliance). All tests must fail (RED state). Standalone or pipeline-aware via env vars. Triggers: test-writer, write tests, TDD, red phase, failing tests.

Use when the user asks to run Gemini CLI (gemini, gemini resume) or references Google Gemini for code analysis, refactoring, or automated editing

Deep architectural review of a platform or product — cross-references code against claims, maps failure modes, evaluates scaling bottlenecks, and produces a decision-grade handoff document with ADRs. Use when: reviewing an existing system for scaling readiness, performing a CTO handoff, evaluating platform architecture for enterprise readiness, or auditing a codebase before a major migration. Triggers: architecture review, scaling review, platform review, CTO handoff, system audit, scaling analysis, architecture assessment, production readiness.

Meta-skill that generates and refines SKILL.md files using the 12 prompting techniques as a quality checklist. The skill that writes skills. Triggers: craft skill, create skill, generate skill, write skill, skill template, meta prompt, improve skill, refine skill.