human-writing-style
Use when generating any written content (essays, legal arguments, discussion posts, docs, reports) to write in Triet's personal voice and style. Triggers include requests like "write this for me", "make this sound like me", "draft an essay", "rewrite in my style", or any content generation where matching Triet's writing patterns matters.
What this skill does
# Human Writing Style
## Overview
This skill produces writing that matches Triet's personal voice, tone, and patterns. It uses an extracted style profile built from 20 real writing samples across essays, legal analysis, advocacy letters, technical documentation, project reports, peer evaluations, science journalism, annotated bibliographies, and discussion posts.
## When to Use This Skill
- Generating any written content (essays, legal arguments, discussion posts, reports)
- Rewriting existing content to match Triet's voice
- User asks for "natural", "human", "my style", or "write like me"
- Content feels generic, robotic, or doesn't sound like Triet after a first pass
## Workflow
### Step 1: Load the Style Profile
Read `triet-writing-profile.md` in this skill's directory. This is the primary reference containing all extracted patterns, signature words, sentence structures, and real examples organized by category.
### Step 2: Identify the Genre
Determine what type of content is being written, then calibrate register:
| Genre | Register | Key Profile Sections |
|-------|----------|---------------------|
| Academic essay | Reflective, opinionated, conversational. Uses "I" freely. | Sections 1, 3, 4, 6, 7 |
| Technical essay | Explanatory, casual grounders ("basically", "pretty complex"). | Sections 4, 5, 6, 12 |
| Legal analysis | Assertive, methodical, evidence-heavy. "We will argue" team voice. | Sections 2, 3, 5, 8, 9 |
| Discussion post / peer response | Warm validation first, then own analysis, then confirmation. | Section 1 (Peer discussion mode), Section 3 (Dual-framework validation) |
| Advocacy / formal letter | Formal, respectful, research-backed. Personal credibility early. | Sections 2, 9, 12 |
| Research / process write-up | Descriptive, chronological, honest about learning curve. | Section 8 (Process narration) |
| Cross-examination / tactical | Direct, parenthetical strategy notes. | Section 2 (Tone by context) |
| Project report / reflection | Sprint-by-sprint narration, lessons-learned, future recommendations. | Sections 8, 12 |
| Peer evaluation | Blunt, evidence-based, grade-justified. No padding. | Sections 1, 12 |
| Product pitch / requirements | Gap identification, feature lists, priority frameworks. | Sections 3, 12 |
| Annotated bibliography | Source-then-explain, "I chose this source because...", synthesis evolution. | Sections 9, 12 |
| Science journalism | "Imagine..." hooks, technical-to-accessible translation, analogy-heavy. | Sections 7, 12 |
| Quiz / exam responses | Conversational-academic blend, concept explanation as if to a peer. | Sections 1, 6 |
| Issue / audience analysis | Controversy framing, audience profiling, explicit rhetorical strategy. | Sections 3, 7, 9, 12 |
| Technical documentation | Feature lists, priority frameworks, technology justifications. | Sections 8, 12 |
| Company report / deliverable | Impersonal or "we" voice (not "I"), findings-first, recommendation-close. Use Triet's transitions and sentence patterns but suppress first-person opinions. | Sections 4, 5, 8, 12 |
### Step 3: Genre-Specific Deep Dive (Optional)
If the profile doesn't provide enough depth for the specific genre, check `source-index.md` in this directory. It maps each genre to original source files that can be read for deeper immersion in that register.
### Step 4: Write Using These Rules
#### Voice
- Default to confident-casual. Opinionated but fair.
- Use "I believe" / "My opinion is" without apology.
- Use "definitely" naturally (Triet's most frequent intensifier).
- Use "Effectively" as a summative transition.
- Use "Given that" as a logical pivot.
#### Structure
- Open paragraphs with context/framing before the argument.
- Follow: Context > Claim > Evidence > Concession > Rebuttal > Reinforcement.
- When using ethical/legal frameworks, apply dual-framework validation (if two frameworks agree, the conclusion is confirmed).
- Close with a strong, concise statement.
#### Sentences
- Favor medium-long compound-complex sentences.
- Use parenthetical asides for precision: "which is one of the most critical tasks".
- End buildup sequences with a short punchy closer.
- Use dash-separated elaboration: "things like X, Y, and Z".
#### Transitions
- Heavy logical connectors: "Furthermore", "However", "Therefore", "In addition to".
- "Effectively" to summarize. "Whereas" for parallel contrast.
- "This then leads to..." for cascading consequences.
- "To reiterate..." / "Essentially, it all comes down to..." for summation.
#### Signature Phrases (Use Naturally, Don't Force)
- "definitely" — sprinkle where emphasis is needed
- "basically" — to ground technical explanations casually
- "pretty [adj]" — casual downtoner in technical contexts ("pretty complex")
- "This directly [verb]..." — for impact claims
- "What's [adjective] here is..." — to highlight key points
- "What makes X [adj] is..." — to emphasize uniqueness
- "The fact is..." — to anchor a core claim
- "rather than" — to reframe alternatives
- "Overall, I..." — for reflective closers
- "Looking back at..." — for retrospective openings
- "In addition with..." — unique quirk (not "In addition to")
#### Authentic Markers (Preserve, Don't "Fix")
- Occasional "Which" fragments after periods
- "meanwhile" as a mid-sentence pivot
- "as to whether" constructions
- "In addition with" instead of "In addition to"
- "softwares" as plural
- "along with" gerund chains ("along with documenting...")
- Stacked positive adjectives in peer contexts ("very interesting and quite relatable")
### Step 5: Self-Audit (Mandatory)
After writing, check EVERY item before outputting. If any fails, fix it.
| Check | Pass? |
|-------|-------|
| Does "definitely" appear at least once (if 500+ words)? | |
| Are there any one-sentence paragraphs? If yes, merge or expand them. | |
| Does every paragraph have 2+ sentences? | |
| Count signature markers — is density below 1 per 100 words? | |
| Scan for banned words and AI anti-patterns (Section 13 of profile). | |
| Does the opening avoid "Let's explore" / "In this essay"? | |
| Does the closing avoid "In summary" / "To wrap up"? | |
| Are there any rhetorical setups like "The question is..." followed by a one-line answer? Remove them. | |
| Read the first sentence of each paragraph — do they sound like press releases? Rewrite. | |
## Anti-Patterns (Never Do This)
These phrases are ABSENT from all 20 of Triet's writing samples. If they appear in output, it doesn't sound like Triet:
- "Let's dive into..." / "Let's explore..." / "Let's break down..."
- "It's worth noting..." / "It is worth noting..."
- "In summary" (use "In conclusion" or "Overall" instead)
- "This is because" as a sentence opener
- "In other words" / "That being said" / "That said" / "On the other hand"
- "Notably" as a sentence opener / "It should be noted"
- One-sentence dramatic paragraphs — Triet never does this. Examples of violations to catch:
- `"The answer is yes."` as its own paragraph
- `"This is the opening."` as its own paragraph
- `"And that changes everything."` as its own paragraph
- **Fix:** Merge into the surrounding paragraph or expand to 2+ sentences.
- Rhetorical question-answer setups: `"The question becomes whether..."` followed by `"The answer is [yes/no]."` — Triet states conclusions directly without theatrical buildup.
- Em-dashes for dramatic pauses (Triet uses dashes for elaboration lists, not drama)
- "Here's what stood out most:" / "Here's the thing:" — generic conversational filler absent from Triet's writing
- Excessive passive voice (Triet defaults to active with "I" and named subjects)
## Paragraph Structure
- Default paragraph: **3-5 sentences** (essays/legal), **2-3 sentences** (discussion posts), **2-4 sentences** (reports/letters)
- One-sentence paragraphs are extremely rare (< 5%). Never for dramatic emphasis.
- Introduction paragraphs tend longer (4-5 sentences). Body paragraphs 3-4.
- Open documents with **broad contextual Related in Image & Video
watch
IncludedWatch a video (URL or local path). Downloads with yt-dlp, extracts auto-scaled frames with ffmpeg, pulls the transcript from captions (or Whisper API fallback), and hands the result to Claude so it can answer questions about what's in the video.
physical-ai-defect-image-generation
IncludedUse when the user wants to orchestrate defect image generation, run associated setup, or handle outputs on OSMO. The Day 0 path handles cold-start with USD-to-ROI, image-edit augmentation, and AnomalyGen to create initial PCBA datasets. The Day 1 path performs inference and labeling on real images. This skill helps with first-time asset setup, creation of finetuning checkpoints, and configuring deployment. Trigger keywords: defect image generation, dig workflow, dig pipeline, defect image detection workflow, aoi pipeline, aoi anomalygen, usd2roi anomalygen, day 0 pcba, day 1 pcba, day 1 real-photo alignment, day 1 manual roi, metal surface anomaly, glass defect, anomalygen finetune, setup_pcb, setup_metal, setup_glass, setup_pretrained, dig setup, dig datasets, dig pretrained checkpoint, dig image-edit endpoint.
accelint-react-best-practices
IncludedReact performance optimization and best practices. ALWAYS use this skill when working with any React code - writing components, hooks, JSX; refactoring; optimizing re-renders, memoization, state management; reviewing for performance; fixing hydration mismatches; debugging infinite re-renders, stale closures, input focus loss, animations restarting; preventing remounting; implementing transitions, lazy initialization, effect dependencies. Even simple React tasks benefit from these patterns. Covers React 19+ (useEffectEvent, Activity, ref props). Triggers - useEffect, useState, useMemo, useCallback, memo, inline components, nested components, components inside components, re-render, performance, hydration, SSR, Next.js, useDeferredValue, combined hooks.
elevenlabs-agents
IncludedBuild conversational AI voice agents with ElevenLabs Platform using React, JavaScript, React Native, or Swift SDKs. Configure agents, tools (client/server/MCP), RAG knowledge bases, multi-voice, and Scribe real-time STT. Use when: building voice chat interfaces, implementing AI phone agents with Twilio, configuring agent workflows or tools, adding RAG knowledge bases, testing with CLI "agents as code", or troubleshooting deprecated @11labs packages, Android audio cutoff, CSP violations, dynamic variables, or WebRTC config. Keywords: ElevenLabs Agents, ElevenLabs voice agents, AI voice agents, conversational AI, @elevenlabs/react, @elevenlabs/client, @elevenlabs/react-native, @elevenlabs/elevenlabs-js, @elevenlabs/agents-cli, elevenlabs SDK, voice AI, TTS, text-to-speech, ASR, speech recognition, turn-taking model, WebRTC voice, WebSocket voice, ElevenLabs conversation, agent system prompt, agent tools, agent knowledge base, RAG voice agents, multi-voice agents, pronunciation dictionary, voice speed control, elevenlabs scribe, @11labs deprecated, Android audio cutoff, CSP violation elevenlabs, dynamic variables elevenlabs, case-sensitive tool names, webhook authentication
humanizer
IncludedHumanize AI-generated text by detecting and removing patterns typical of LLM output. Rewrites text to sound natural, specific, and human. Uses 28 pattern detectors, 560+ AI vocabulary terms across 3 tiers, and statistical analysis (burstiness, type-token ratio, readability) for comprehensive detection. Use when asked to humanize text, de-AI writing, make content sound more natural/human, review writing for AI patterns, score text for AI detection, or improve AI-generated drafts. Covers content, language, style, communication, and filler categories.
generating-mermaid-diagrams
IncludedSalesforce architecture diagrams using Mermaid with ASCII fallback. Use this skill when generating text-based diagrams for Salesforce architecture, OAuth flows, ERDs, integration sequences, or Agentforce structure. TRIGGER when: user says "diagram", "visualize", "ERD", or asks for sequence diagrams, flowcharts, class diagrams, or architecture visualizations in Mermaid. DO NOT TRIGGER when: user wants PNG/SVG image output (use generating-visual-diagrams), or asks about non-Salesforce systems.