tap-skills

Publish docs (plans, research, shapes, alignment atlases) to the private Dossier platform at teambrilliant.dev, pull them back with comments on any machine, and manage external share links. Use when someone says "publish to dossier", "dossier publish", "publish this doc/plan", "push this to teambrilliant.dev", "share this doc with the team", "share this externally", "pull the plan from dossier", "get the comments on the doc", or when resuming work on a plan that lives on Dossier. Scope guard - "publish" alone is ambiguous (npm, blog, git); this skill is only for the Dossier doc platform. Requires DOSSIER_TOKEN in the environment. Pairs with tap-skills:render-doc, which produces the HTML this skill uploads.

Generate and maintain an alignment-diagram atlas — a self-contained file:// viewer in a target directory (default `.tap/diagrams/atlas/`; supports several atlases per repo, e.g. per PARA area) whose tiles open per-flow alignment grids (layers × steps, line of visibility, moments of truth, coverage honesty). Use when someone says "alignment diagram", "alignment map", "experience atlas", "map this flow", "diagram how this works end to end", "service blueprint", "Kalbach map", or wants one navigable surface showing how a product behaves / an engine runs, flow by flow. Auto-detects scaffold / map-a-flow / review mode. Outputs to `.tap/diagrams/atlas/`.

Curate and maintain `.tap/product.md` — a compressed, agent-readable product-context file (what we build, audience, current focus, bets, non-goals). Use when someone says "curate product context", "install product context", "update product context", "product vision in repo", "what are we building", or when an agent enters a repo and can't find durable product-strategic context. Auto-detects interview / review / refresh mode. Outputs to `.tap/product.md`.

Explore a running web app in a browser and build or update `.tap/smoke-tests.md` — a lean catalog of post-release smoke-test cases. Use whenever someone says "build smoke tests", "create a smoke test catalog", "explore the site for testing", "what should we smoke test", "update the smoke test catalog", "add smoke test cases", or wants exploratory testing of a web app to capture what matters before a release. Use it proactively when entering an app that has no `.tap/smoke-tests.md` and the user wants release coverage. Pairs with `qa-smoke-run`, which executes the catalog this skill produces. Outputs to `.tap/smoke-tests.md`.

Deterministically convert a markdown doc (plan, research, shape, report) into one self-contained HTML file with house styling, TOC, checklists, and mermaid support. Use when someone says "render this doc", "render this plan", "make this markdown a web page", "convert md to html", "make this plan shareable as HTML", or before publishing a markdown doc with tap-skills:dossier-publish. NOT for authoring content — the markdown is the source of truth; this only renders it.

Just-in-time retrospective focused on improving agent autonomy. Use when someone says "retro", "retrospective", "what did we learn", "what went wrong", "post-mortem", "incident review", or after a feature ships, an incident resolves, a pattern of agent failures emerges, or any event worth reflecting on. Analyzes what happened, identifies what blocked agent autonomy, and produces concrete improvements (learnings + tickets). Not calendar-driven — event-driven. The learning loop that makes the system self-improving.

Measure the health of a software development system using stocks, flows, and feedback loops. Use when someone says "systems health", "how's the project going", "health check", "measure our process", "are we shipping fast enough", "what's slowing us down", or for periodic check-ins on development velocity and quality. Pulls data from git, GitHub, and CI to diagnose what's working and what's broken. Outputs to .tap/system-health.md.

Assess how ready a repository is for autonomous agent work. Use when someone says "audit this repo", "tap audit", "how ready is this codebase", "assess this project", or when an agent enters an unfamiliar codebase and needs to understand it before working. Scans documentation, MCP servers, CLI tools, permissions, test infrastructure, environments, and process to produce a readiness assessment with actionable leverage points. Outputs to .tap/tap-audit.md.

Build a 12-month outcome-based technology roadmap that justifies budget to CEO and board. Use when someone says "tech roadmap", "technology roadmap", "build a roadmap", "justify budget", "board presentation", "budget justification", "12-month plan", "annual technology plan", or when a CTO needs to structure technology investments around business outcomes. Guides the CTO through assessing current state, defining strategic focus areas, mapping outcomes to investments, and producing a board-ready markdown document. Outputs to .tap/tech-roadmap.md.

Assess what's needed to make feedback loops autonomous in a repo. Use when someone says "loop check", "what do I need to work autonomously", "check my feedback loops", "what's manual here", "what should I automate", "can an agent iterate here", or before starting work in an unfamiliar repo to understand what's missing for autonomous iteration. Also use when the user asks "what do you need to make this autonomous?" or describes a workflow they want to close the loop on. NOT for: full repo audits (use tap-audit), coding, test writing, or implementation.

Harvest course-corrections from the current conversation and convert them into durable fixes so the agent doesn't need the same steer next time. Use when someone says "tighten the loop", "tighten loop", "debrief this session", "session debrief", "what should I update so next time you don't…", "how can I make your life easier", "what tripped you up", "what slowed you down", or at the end of a session when the user reflects on agent friction. The scope marker is THIS SESSION / THIS CONVERSATION — that's the discriminator from sibling skills. Reads the transcript only, classifies each steer, routes to the right fix tool. NOT for: assessing repo readiness before work (use loop-check), retros on past PRs/incidents/releases (use tap-skills:retrospective), coding, or applying edits inline.

Execute the smoke-test catalog at `.tap/smoke-tests.md` against a running web app in a browser, then report only what's broken. Use whenever someone says "run smoke tests", "run the smoke test catalog", "execute smoke tests", "smoke test the release", "qa smoke run", or wants to verify a release or deploy against the app before shipping. This is the per-release regression run. Pairs with `qa-smoke-catalog`, which authors the catalog this skill consumes. Reports pass/fail in chat; saves failure artifacts to `.tap/qa-runs/<timestamp>/`.

Analyze the impact surface of a PR or set of changes before merging. Use when someone says "blast radius", "review this PR", "what does this change affect", "is this safe to merge", "impact analysis", or after an agent opens a PR that needs human verification. Maps what changed, what else is affected, what could break, assigns risk level, and generates a manual verification checklist. The human gate for mixed human-agent teams.