vite-flare-starter

Evaluate multiple choices side-by-side based on relevant criteria, then present a comparison the user can act on. Use when the user asks to compare options, choose between alternatives, or weigh up X vs Y.

Day-of-week rotating proactive sweep — Mon=connections / Tue=routines / Wed=skills / Thu=memories / Fri=findings / Sat=cross-cutting / Sun=weekly summary. Forces outward scan instead of inward thrash. Routine-only; load this skill into a daily routine that fires AssistantAgent. Adapted from goanna's caretaker.

Use when an agent has had N consecutive quiet runs with nothing to do. Pivots to bounded inward consolidation (promote findings, refresh memory, audit asks) anchored in own data — NOT generic news summarisation. Adapted from goanna's reverie cycle. Routine-only; loaded by the routine that detects the quiet condition.

Use to track open questions you owe the user (asks) and time-bounded commitments you owe yourself (tasks). Write OPEN entries when you commit; promote to CLOSED when answered/done. Survives session compaction. Adapted from goanna's asks.md/tasks.md pattern.

Reviewer criteria for AI-generated code via with_review. Checks for security issues (injection, auth bypass, secret leakage), correctness, and project conventions. Use as criteria.skill when generating code that touches data, auth, or external systems.

Reviewer criteria for outbound email drafts via with_review. Checks tone-recipient match, professionalism, no AI tells, no placeholder leakage. Use as criteria.skill argument when reviewing email drafts before send.

Reviewer criteria for AI-generated summaries via with_review. Checks faithfulness to source, no hallucinated facts, appropriate length, key points preserved. Use as criteria.skill when summarising articles, documents, transcripts, or batch-task outputs.

Default reviewer criteria for the with_review tool. Generic quality bar — accuracy, intent-match, no hallucinations, clarity. Use as the criteria.skill argument when calling with_review without a domain-specific reviewer skill.

SessionEnd hook for routines — turns raw error messages into actionable findings (e.g. "Error: 401" → "Gmail token expired — reconnect at /dashboard/connectors"). Routine-only.

Weekly cross-agent curation routine — surfaces patterns across agents, promotes stable ones into shared knowledge, writes a weekly digest into Inbox. Routine-only.

End-of-day distillation routine — reads recent runs + findings, decides what graduates / dismisses / stays open, writes a one-paragraph day summary. Routine-only.

SessionEnd hook for routines emitting findings — picks the right channel (inbox_add / notify / approval_queue / space_send / webhook_post) and writes a short run-summary line. Routine-only.

Daily meta-routine — watches every other routine for error rates, drift, and runaway cost; surfaces routine_health findings into the Inbox. Routine-only.

Calibrates finding importance (high / medium / low) before inbox_add — used by routines for consistent thresholds. Routine-only.

Analyse a CSV file in the sandbox — row/column counts, column types, summary statistics for numeric columns, unique value distributions for categorical columns, and missing-value counts. Use when the user asks to explore, summarise, profile, or analyse a CSV file.

Summarise a git diff into a short change description, risk assessment, and suggested commit message. Use when the user asks to review changes, draft a commit message from a diff, or wants a high-level overview of what was edited.

Turn a meeting audio recording into structured notes — attendees, agenda items discussed, key decisions, action items with owners, and open questions. Save the result as a Markdown file. Use when the user attaches an audio file of a meeting or explicitly asks for meeting notes from audio.

Research a topic by searching the web, fetching the most relevant pages as markdown, and producing a well-structured summary with citations. Use when the user asks to research, investigate, find information about, or look into a topic.

Create a polished visual report as an HTML artifact with charts, tables, and formatted text. Use when the user asks for a report, dashboard, analysis, or visual summary of data.

Review code for bugs, security issues, performance problems, and style. Use when the user pastes code, shares a file, or asks "what's wrong with this code?".

Answer questions about an uploaded document (PDF, image, or text file). Use when the user has attached a file and asks something specific about it.

Write a professional email tailored to a specific recipient, purpose, and tone. Use when the user asks to draft, write, or compose an email.

Extract structured data (JSON) from messy text, documents, or web pages. Use when the user asks to extract, parse, structure, or pull data out of unstructured content.

Verify a specific claim by finding multiple authoritative sources that confirm or contradict it. Use when the user asks to fact-check, verify, or check whether something is true.

Generate a personalised morning briefing covering weather, news, schedule, and saved priorities. Use when the user asks for a morning brief, daily summary, or "what's happening today?".

Break a goal into a clear sequence of actionable steps with dependencies and estimated effort. Use when the user describes something they want to accomplish and asks for a plan, breakdown, or "how do I...".

Review the current conversation and save key facts to long-term memory. Use when the user asks you to remember things from this chat, save important info, or "remember this for next time".

Rewrite text for a different audience, tone, length, or format while preserving the core meaning. Use when the user asks to rewrite, rephrase, simplify, formalise, or adapt text for a specific audience.

Save a piece of research, notes, or content to the user's filesystem as a well-organised markdown document with metadata. Use when the user asks to save, archive, file away, or "keep this for later".

Fetch a URL and produce a structured summary with title, key points, and an action-oriented takeaway. Use when the user shares a link and asks for a summary, TL;DR, or "what's this about?".