synthesis-skills

Coordinate an absence end to end — vacation, conference travel, family visits, medical leave — the way a good chief of staff would. Covers the notification order that protects relationships (principals hear it from you, never from their own assistant or a group channel), coverage and reachability as required content rather than afterthoughts, recurring-meeting release, the personal-continuity tier that keeps a trainer or therapist or tutor in the loop while you travel, travel-logistics forwarding, out-of-office set AND clear, and the return sweep. All names, tiers, channels, and lead times load from a private config, so the skill is publishable and the configuration is yours. Use when planning time off, booking work travel, announcing an absence, arranging coverage, or returning from one.

Run bounded, differently-shaped-agent adversarial review against the principal's outcome, with artifact-complete rounds, production-topology handoffs, explicit concessions, a fail-closed finding ledger, sufficiency rulings, and independent post-publication acceptance. Use for adversarial review, cross-agent review, red-team collaboration, review rounds, finding-ledger work, or reviewer handoffs.

Audit, install, and verify a synthesis ecosystem across multiple AI agent runtimes. Use for Claude Code and OpenAI Codex parity audits, AGENTS.md and CLAUDE.md instruction migrations, skill or plugin deployment checks, lifecycle-hook health, Mac bootstrap validation, active-project handoffs, post-compaction recovery, and any request to make synthesis project management portable between agent clients.

Compose and send honest agent correspondence across Slack, email, and other channels. Defines principal-direct, assistant, and bot lanes; the voice axis (chief-of-staff personas speak as the principal, executive-assistant personas speak as themselves); review-depth governance; persona configuration; disclosure signatures; and compose/send gates. Use for agent correspondence, sending on a principal's behalf, message signatures, disclosure lanes, persona registries, agent branding, agent voice, third-person agent messages, standing-direction sends, ghostwriting disclosure, or outbound-message gates.

Discipline for catching the lazy-shortcut antipattern in AI-assistant output. Makes the costume vocabulary explicit so agents recognize when their drafts have slid into deferral, dismissal, or false consultation. Includes the constraint-first protocol, sub-agent dispatch and acceptance hygiene, and a pre-response self-check. Use when asked to: avoid shortcuts, audit for laziness, check for deferral, enforce best solution, no shortcuts, anti-shortcut, constraint-first, sub-agent hygiene, costume vocabulary.

Refresh and revitalize older articles with improved clarity, presentation, and completeness while preserving the author's voice, argument, and period-appropriate context. Standard editorial practice for keeping a published archive polished and accessible. Use when asked to: refresh article, revitalize article, update older article, polish published post, enhance older content.

Five-phase workflow for creating high-quality thought leadership articles: reader briefing, research and validation, strategic writing, pre-publication critical review, and publication-package review (title, metadata, and batch gates). Includes anonymization protocol, credibility assessment, substance density checks, engagement optimization, the title-only stranger test, the title/description/body truth contract, batch headline monotony budgets, and the slug/metadata closure invariant. Use when asked to: write article, thought leadership, blog post, article workflow, write blog, draft article, create thought piece, write opinion piece, leadership article, headline review, title review, publication readiness, article package review.

Execute an explicitly delegated whole task autonomously using the thinking framework, durable plan and context, checkpoints, anti-shortcut discipline, and implementation-integrity gate — and, for unattended runs, a verified continuation mechanism with budget and runaway control, so overnight and multi-day engagements keep producing turns instead of idling silently. Activate only for clear end-to-end delegation such as 'autopilot this,' 'take care of this for me,' 'handle this end to end,' 'run overnight,' or 'complete all phases autonomously'; never infer it from a single-step approval, discussion of autonomy, or ambiguous wording.

Work with Bitbucket Cloud repos through the open-source bkt CLI — PR lifecycle (list, view, diff, comment, approve, merge), repo/branch reads, and the bkt api escape hatch. Encodes auth setup, the context-host gotcha, a gh-to-bkt command map, and write-safety rules so agents use one consistent command surface instead of reinventing REST calls. Use when asked to: bitbucket, bkt, bitbucket pr, open bitbucket pr, review bitbucket pr, bitbucket cli, bitbucket api.

Reconcile pending, missed, and incomplete commitments after any gap in the daily-ritual cadence (travel, family visits, crunch weeks). Sweeps daily plans, transcripts, and project context over an arbitrary window; classifies every surfaced item (including expired-for-learning); produces a dated catch-up ledger document; routes survivors without flooding the daily plan. Use when asked to: catch up on missed tasks, sweep pending items, catch-up ledger, reconcile backlog, what did I miss, post-vacation catch-up, falling behind recovery.

Mid-session refresh and drift-recovery protocol. Re-syncs the agent's awareness of current date, project state, recent commit history, and concurrent root-session claims. Use when asked to: checkpoint, re-sync, refresh context, mid-session refresh, recover from compaction, re-read project state, verify dates, drift check, coordination check, where are we, what was last done, am I caught up. Invoke on any suspicion of context drift, after long conversation pauses, before generating time-interval claims, or before writes when another agent session is active.

Act as a principal's chief of staff and executive assistant. Protects the principal's time through meeting triage, calendar-aware scheduling, look-ahead reviews, overcommitment checks, tracked holds, travel planning, correspondence posture, and a follow-up ledger; personal rules load from private preferences. Use for scheduling, meeting requests, calendar-related replies, calendar defense, travel, or any chief-of-staff and executive-assistant duty.

Enforce clean-text and no-hidden-marker requirements, audit inspectable characters and provenance, and state the verification boundary for statistical text marks. Use when generating clean text, checking hidden characters, addressing watermark concerns, or selecting a controlled generation path.

Systematic 10-dimension quality scan of code diffs, producing scored PASS/WARNING/FAIL verdicts per dimension with a machine-readable overall result. Includes PR review mode for cross-referencing findings against existing reviewer comments. Use when asked to: code audit, audit my changes, quality check, review the diff, check this code, audit my code, scan for issues, code quality scan, diff review.

The adopt-and-adapt integration pattern for multi-contributor synthesis-coded projects. Covers the lead synthesist role, quality gates, cherry-pick safety, and graduated integration intensity. Use when asked to: synthesis coding, multi-contributor, integration, cherry-pick, adopt and adapt, lead synthesist, integrate contributions, merge contributor work.

Structured approach to code generation, implementing features, and writing code. Use when asked to generate code, implement a feature, write code, or tackle a coding task. Analyzes the task, generates multiple approaches with trade-offs, selects the optimal solution, and implements it.

Enterprise-scale codebase audit methodology with tiered review system (Essential through Mission-Critical). Use when asked to: codebase review, code audit, code review, review codebase, architecture review, security audit, full code review, enterprise review, codebase health check.

Condense messages to 5 sentences or less using the High-Five Habit framework. Produces concise, effective business communications across email, Slack, text, and social media. Use when asked to: condense message, shorten email, brief message, concise writing, email template, make shorter, tighten email, reduce message length, write concise message.

Strategically promote and distribute content across social media platforms to build thought leadership, spark discussion, and create professional connections. Provides platform-specific guidelines, engagement strategies, cross-platform coordination, and quick-start templates for blog promotion and event/topic posts. Use when asked to: promote content, content distribution, social media strategy, share article, social media post, write tweet, LinkedIn post, Instagram post, promote blog, blog promotion, post about event, content marketing, distribute post, promote article, social media promotion.

Content framing methodology for synthesis engineering articles. Covers topic selection, sophistication standards, engagement patterns, quality gates, and operational rules for publishing technical thought leadership. Use when writing synthesis engineering content, framing technical articles, checking content quality gates, or publishing thought leadership.

The most comprehensive open-source slop detection system available. Catches AI-generation patterns by model family (Claude, GPT, Gemini, Llama, Grok, DeepSeek, Mistral, Qwen), substance and depth failures (beautiful word salad), and the full v3.1.0 catalog refreshed. Includes causal-layer attribution, combined-signal fingerprints, two-axis calibration, ESL safe-harbor, and zone-conditional detection (artifact mode vs full-response mode). Spans the entire LLM era through the compounding-archive principle. Use for content quality, slop detection, AI content auditing, editorial review, content improvement, and publishing standards.

Three-tier context architecture for managing AI working memory across long-running projects. Use when asked to: manage context, project context, session management, context lifecycle, working memory, archival, archive sessions, context maintenance, garbage collection for context, tiered context.

Configure an LLM as a creative writing coach and editor. Use for creative writing, writer setup, configure writer, creative assistant, content creation coaching, storytelling assistance, and writing improvement.

Day-start and day-end checklists for synthesis engineering projects. Execute dependency-ordered rituals for context optimization, channel sync across Slack, Google Chat, email, meeting transcripts, and document comments, catch-up reads, PR reviews, day planning, and communications. Use when asked about: daily ritual, morning routine, day start, day end, daily checklist, morning checklist, end of day checklist, daily workflow.

Collect many parallel decisions from a principal in one sitting instead of one per turn. Generates a self-contained HTML packet — one row per decision carrying the item, the agent's recommendation, the reasoning, and a link — with buttons, a per-row note box, local persistence, and a paste-able summary the principal returns in a single message. Use when you owe five or more decisions of the same shape; when a review, migration, upgrade, triage, or backlog pass has produced a list someone must rule on; or when a per-item conversation is burning round-trips.

Two-category disclosure governance for people who publish under their own name while handling confidential work. Distinguishes published-precedent facts (deliberately public biography an agent may restate) from unapproved disclosures (anything learned from private context), with a precedent ledger, surface classes, five decision tests, and git-hook enforcement. Use when asked about: disclosure policy, confidentiality guardrails, can I name this company, precedent ledger, public bio names, publication surface, unapproved disclosure, name allowlist.

Translate technical work into clear, actionable communication for CEOs, CFOs, business-unit leaders, boards, and other non-technical executives. Applies persona tests, de-jargoning, mechanism-to-consequence translation, and trusted upward-report structure. Use for executive updates, progress reports, executive summaries, board updates, leadership status, business-term translations, or any technical message to a business audience.

Systematic fact-checking for articles, blog posts, news content, and AI-synthesized material. v2.0 adds nine new protocol sections covering nested attribution, paraphrase drift, composite quotes, position-shifting, source-translation drift, URL rot vs hallucination, AI-generated synthetic sources, citation laundering chains, and tool-specific hallucination patterns by LLM family. Use when asked to: fact-check, verify claims, verify sources, check accuracy, citation verification, review factual accuracy, validate references, audit AI-summarized content.

Deterministic pre-commit policy for the synthesis-engineering workflow. Enforces staged paths against an active session's lease-backed coordination claim when a board is configured. Classifies each repo by publication surface (personal / public-surface / strict) from its push remotes, applies a tiered pattern set: Tier 0 credentials always; Tier 1 financial / HR / confidentiality / client names in strict and public-surface repos — public-surface minus only the disclosure ledger's published-precedent names. YAML-driven policy lives in ~/.synthesis/git-hook-config.yaml. Use when asked to: install git hooks, configure pre-commit policy, enforce coordination claims, prevent credential leaks, prevent confidential-name leaks, allow published bio names on my own sites, disclosure ledger enforcement, set up the synthesis-engineering enforcement layer.

Evidence and provenance discipline for AI-agent output — the truth-side companion to synthesis-anti-shortcuts' effort-side discipline. A catalog of grounding rules: never record imagined events, quote only what a tool surfaced, treat context files and memories as caches to re-verify before propagating, name the layer a config claim describes, read the evidence in hand before theorizing, count the corpus before generalizing, prove absence with a positive control and bounded reads, never complete truncated output, and validate paths before writes and deletions. Use when asked to: grounding discipline, verify claims, evidence check, provenance check, anti-confabulation, absence claim, negative finding, zero results, cache vs truth, count the files, truncated output, verify paths, safe deletion, is this grounded.

Post-implementation verification protocol that catches incomplete work, hidden shortcuts, and gaps between 'tests pass' and 'production works.' Systematically traces data chains, detects placeholders, audits test honesty, and verifies environment parity. Use when asked to: verify implementation, check completeness, implementation review, is this done, built properly, no shortcuts, verify work, completion check, integrity check, self-review, pre-PR check, ship check.

Manifest-driven email inbox cleanup across three tool stacks: iCloud / generic IMAP via Python + YAML rules; Microsoft 365 + outlook.com via Mail.app AppleScript; Gmail via workspace-mcp Gmail API and server-side filters. Engine is public; per-user rules live privately at ~/.synthesis/inbox-cleanup/. Ships with prompt-injection defenses (sanitization module + adversarial test fixtures) for any LLM-augmented path. Use when asked to: clean up inbox, sweep email, categorize senders, build email rules, archive promotions, set up Gmail filters, build email cleanup automation.

Edit, validate, and ship Markdown knowledge-base changes through a repository's config-driven workflow. Reads .agents/knowledge-base.yaml for the editable bundle, generated/refused paths, topic routing, frontmatter schema, confidentiality control, Git host, branching, and review policy. Use when a user asks to update a knowledge base, edit KB content, fix a durable fact, add a concept, ship an existing KB edit, open a knowledge-base review request, or synchronize a local knowledge-base checkout after publication.

Capture durable session facts in an OKF knowledge base with corpus-wide deduplication, confidentiality routing, conflict reconciliation, provenance, validation, and repository-aware shipping through synthesis-kb-edit. Use at session end, when asked to capture or update knowledge, merge facts into ai-knowledge, or preserve corrected facts about people, organizations, products, decisions, or strategy.

Find authoritative hyperlinks for people, organizations, and entities mentioned in content; recover dead URLs in archives via Wayback / Wikipedia / strip prioritization. Use when asked to: find links, hyperlink, research links, find URLs, gather references, look up websites for people or organizations, recover dead links, fix link rot, find archived versions.

Guide for setting up and configuring LLM projects across platforms. Use when setting up a Claude Project, creating a ChatGPT GPT, configuring a Gemini Gem, setting up an AI assistant with custom instructions and knowledge, or configuring any LLM platform with AI Knowledge content.

Profile a computer, compare Ollama, LM Studio, llama.cpp, and MLX-LM, recommend local open-weight model artifacts that fit its real memory and storage, install approved artifacts through deterministic managed-runtime adapters, update installed Ollama models with before-and-after identity receipts, maintain a privacy-safe per-machine inventory, and verify local inference. Use for: local models, open weights, Ollama, LM Studio, llama.cpp, MLX model selection, model updates, which model fits this Mac or PC, install Qwen/GLM/Kimi/DeepSeek locally, hardware profile for LLMs, model inventory, local inference benchmark.

Multi-Mac configuration sync via iCloud with bidirectional config file sync, git repository sync, machine inventory, and one-time action system. Use when asked to: mac sync, sync config, sync repos, sync with GitHub, push config to iCloud, pull config from iCloud, run mac-sync, repo status.

Fetch AI-generated meeting notes and full transcripts into local markdown files, and verify that an artifact is primary before it supports attribution-bearing claims. Tool-agnostic: works with Gmail/Drive MCPs and local transcript exports. Use when asked to fetch, pull, sync, download, import, classify, or verify a meeting transcript.

Fail-closed pre-send enforcement for agent-drafted correspondence. A PreToolUse hook blocks every message-sending or draft-creating tool call unless the outgoing text passes a deterministic register scan AND a fresh, single-use grounding ledger — sha256-bound to the exact message — attests that the composing agent read the full thread, searched prior correspondence, and mapped every factual claim to a source. Use when setting up, debugging, or composing under the guard; when a send is blocked; or when asked about message grounding, voice enforcement, or pre-send gates.

Cross-provider model-tier convention for agentic work: three role labels (judgment, routine, bulk — formerly frontier, efficient, light) resolved to current model IDs per provider in tiers.yaml, so skills, project docs, and memory never hardcode model names. Also carries the role-selection rule: route by whether the CAUSE is known, not by how small the task looks — a symptom report is diagnosis and belongs in judgment even when the subject is one file. Use when asked about: model tiers, which model, model selection, judgment model, routine model, bulk model, frontier model, efficient model, switch models, model equivalents across providers, update model table, which effort level, low effort, wrong model for the task.

Validate, convert, and author content for Google's Open Knowledge Format (OKF v0.1) — the markdown-plus-YAML-frontmatter knowledge-bundle spec announced 2026-06-12. Includes a conformance validator (the checker Google's own OKF repo ships none of), a converter that backfills OKF frontmatter onto an existing markdown corpus idempotently, a config-driven seven-point metadata-consistency checker, and the proven repo-by-repo conversion procedure. Use when adopting OKF for an "LLM wiki" style knowledge base, auditing conformance, checking frontmatter/body drift or taxonomy consistency, or converting a corpus of markdown notes/docs into a conformant, agent-readable bundle.

Install, update, repair, diagnose, and uninstall the synthesis work system through one stable public CLI. Covers immutable release acquisition, stable and edge channels, exact pins, full and skills-only profiles, tracked dual-client workspace instructions, optional declarative organization enrollment, transactional desired and observed state, plugin currency, and outcome verification. Use when asked to onboard, install synthesis, set up the ecosystem, configure a knowledge workspace, update or repair an installation, enroll an organization, verify an install, or diagnose onboarding.

Delta review methodology for pull requests in synthesis-coded projects, covering regression risk assessment, root cause analysis, and integration with the adopt-and-adapt workflow. Use when asked to: PR review, pull request, code review, review PR, delta review, review pull request, check PR, evaluate PR.

Pre-merge quality gate framework with six orthogonal dimensions: branch hygiene, clean tree, tests and types, code audit, temporary considerations, and commit history. Produces a mechanical go/no-go verdict. Use when asked to: preflight, pre-merge check, ready to merge, can I ship this, branch ready, quality gate, merge readiness, pre-PR check.

Architecture-decision pre-planning for tickets or issues with real design choices. Runs a structured Q&A loop that locks the load-bearing architectural decisions before any commit plan is drafted, then hands a clean, reviewable decision set to your planning step. Use when asked to: preplan, pre-plan this ticket, let's pre-plan, lock decisions for, design questions for, what are the open questions on, plan a ticket with real design choices.

Lightweight project management system designed for human-agent collaboration, optimized for context preservation and cross-agent coordination across sessions. Use when asked to: project management, project setup, project tracking, synthesis project, manage project, set up project, project structure, session protocol, parallel root sessions, advisory locks, cross-agent coordination.

Configure and run a fail-closed publication promotion gate that builds in isolation, derives output routes from frontmatter, consumes identity-bound representations from the destination parser or renderer, binds receipts to exact inputs and renderer surfaces, and permits a state-changing promotion command only after immediate revalidation. Use for publication gates, outward-surface cleanliness, rendered-output inspection, publishable-range contracts, or promotion receipts.

Stand up and operate a low-cost, read-mostly companion session for ad hoc workspace lookups without pulling a focused project session off task. Every answer carries its source and a Verified, Cached, or Uncertain confidence tier. The workspace's tracked instruction source routes future sessions to the companion automatically. Bootstraps a missing personal knowledge workspace through the public synthesis CLI. Use for an FAQ assistant, quick-answers session, lookup companion, ask-me-anything session, fast Q&A project, or one-off lookup that is not the focused project's task. Not for deep project work, decisions, drafting, sending messages, or work that belongs in its own project.

Pre-writing reader briefing methodology for public articles. A required precondition for any article that draws on internal source material (lessons, project context, codebases) and is published to an external audience. Catches insider context collapse before it becomes a draft, and declares the article's series-dependency state (standalone, standalone after compact context, or true prerequisite). Use when asked to: write article, plan article, brief audience, audience analysis, reader briefing, pre-writing, before drafting, who is this for, series dependency, prerequisite article.