@monoes/monomindcli 2.15.2 → 2.15.4

package/org-skills/avoid-ai-writing/references/pattern-catalog.md

@@ -0,0 +1,126 @@
+# Pattern catalog
+
+Every entry lists the tell and the fix. Quoted strings are illustrative examples, not text to preserve.
+
+## Severity
+
+Every pattern below falls into one of three tiers.
+
+- **P0, credibility killers:** cutoff disclaimers ("As of my last update"), chatbot artifacts ("I hope this helps!"), vague attributions ("Experts believe"), significance inflation on routine events, unfilled placeholders (`[Your Name]`), chat citation markup (`citeturn0search0`), AI-tool tracking parameters (`utm_source=chatgpt.com`).
+- **P1, obvious AI smell:** Tier 1 vocabulary, template and slot-fill phrases, "Let's" openers, synonym cycling, formulaic openings, bold overuse, em dashes above one per 1,000 words, future-narrative closers, hedge stacks ("could potentially"), moral adjectives on non-agentic nouns ("an honest shape"), bullet lists of bare noun phrases.
+- **P2, polish:** generic conclusions, compulsive rule of three, uniform paragraph length, copula avoidance ("serves as", "boasts"), transition phrases ("Moreover", "Furthermore"), Tier 3 repetition.
+
+## Formatting
+
+- **Em dashes:** target zero, hard max one per 1,000 words, headings included. Catch both the Unicode em dash and the double-hyphen substitute. Carve-out: a dash separating a bolded lead term or a link from its gloss inside a list item (`- **Term** - description`) is typography, not a prose splice. A mid-sentence splice still counts.
+- **Bold overuse:** one bolded phrase per major section at most. If something deserves bold, restructure the sentence to lead with it.
+- **Emoji in headers:** remove. Social posts may carry one or two at the end of a line.
+- **Excessive bullets:** convert bullet-heavy sections to prose. Keep bullets for genuinely list-shaped content: feature comparisons, ordered steps, API parameters.
+- **Curly quotes in plain-text contexts:** a weak paste-from-chat signal, meaningful only where nothing auto-curls (code comments, commit messages, plaintext drafts). Word, Google Docs, macOS, and iOS curl by default, so most human prose has them too. Never conclusive. Do not flag curly apostrophes alone.
+- **Immaculate typography in casual registers:** perfect spacing and capitalization in issue comments, chat, or DMs is corroborating evidence at most. The inverse matters more: when editing a human's casual text, keep their typos and idiosyncratic capitalization rather than smoothing them away.
+- **Inline-header lists:** bullets whose bold header repeats the first words of the item ("**Performance:** Performance improved by..."). Strip the header and write the point.
+- **List-label periods:** LLMs end a bullet's short label with a period, then run the gloss as a separate sentence (`- **Intros.** Years of conferences...`). A person writes `- **Intros:** years of conferences...`. Carve-out: a label span that is a full sentence keeps its period.
+- **Title case headings:** use sentence case for subheadings.
+- **Hyphenated-pair overuse:** two problems. First, strings of compound modifiers piled on one noun; cut to the one that matters. Second, the predicate error. Hyphenate before the noun ("a high-quality report") and not after a linking verb ("the report is high quality").
+- **Excessive structure:** more than three headings in under 300 words, or eight-plus bullets in under 200 words, is scaffolding standing in for content. Formulaic headers ("Overview", "Key Points", "Summary") say nothing. A heading followed by a one-line warm-up that restates it should lose the warm-up.
+
+## Sentence structure
+
+- **"It's not X, it's Y."** Rewrite as a direct positive claim. Includes the split-sentence form across two sentences, the multi-negation countdown ("Not the price. Not the features. The trust."), and the tailing negation fragment ("...comes from the selected item, no guessing"). Carve-out: spec constraints in a list ("no dependencies, no telemetry").
+- **Hollow intensifiers:** cut `genuine`, `truly`, `quite frankly`, `to be honest`, `let's be clear`, `it's worth noting that`.
+- **Vague endorsement:** `worth reading`, `worth a look`, `worth your time` substitute a thumbs-up for a reason. Say why.
+- **Hedging:** cut `perhaps`, `could potentially`, `it's important to note that`.
+- **Hedge-stacked predictions:** a modal plus a hedge adverb: "could potentially create", "may eventually unlock". Either word alone is fine; the stack asserts nothing. Pick one.
+- **Missing bridge sentences:** if paragraphs could be reordered without the reader noticing, add connective tissue.
+- **Compulsive rule of three:** vary groupings. One "adjective, adjective, and adjective" per piece at most.
+- **Copula avoidance:** "Serves as", "features", "boasts", "presents", "represents" instead of "is" and "has". Default to the plain verb.
+- **Subjectless fragments and agentless passives:** "No configuration file needed." "Support for nested queries was added." Name the actor when it clarifies. Carve-out: terse reference registers where the fragment is correct, such as changelogs, parameter docs, and commit subjects.
+- **Synonym cycling:** "Developers... engineers... practitioners... builders" in one paragraph. Repeat the clearest word instead.
+- **Parenthetical hedging:** "(or, more precisely, Y)", "(and perhaps more importantly, W)". Give the aside its own sentence or cut it.
+- **False concession:** "While X is impressive, Y remains a challenge." Both halves are vague. Name the specific thing or pick a side.
+- **Invented contrast-pair mirroring:** one half of the contrast is a real term of art, the other is fabricated for symmetry: "false precision rather than genuine accuracy". Reach for a real opposite or state the positive claim.
+- **False ranges:** "From the Big Bang to dark matter." Sweeping breadth that names nothing. List the actual topics.
+
+## Openers, transitions, closers
+
+- **Formulaic openings:** broad context before the point ("In the rapidly evolving world of..."). Lead with the news.
+- **Transition phrases:** "Moreover", "Furthermore", "Additionally", "In today's X", "When it comes to", "At the end of the day", "In conclusion". Restructure so the connection is obvious.
+- **Reader-steering frames:** "Here's what's interesting", "Here's what caught my eye". Let the content signal its own importance.
+- **"Let's" constructions:** "Let's explore", "Let's break this down". Start with the point.
+- **Rhetorical question openers:** "But what does this mean for developers?" If you know the answer, say it.
+- **Speculative scenario openers:** "Imagine a world where...". The scenario does the persuading and no evidence arrives. Carve-out: fiction, a thought experiment with a stated payoff, and instructional "imagine you have a sorted array".
+- **Numbered list inflation:** "Three key takeaways", "Five things to know". Only number a list when the content genuinely has that many parallel items.
+- **Generic conclusions:** "The future looks bright", "Only time will tell", "As we move forward". Cut.
+- **Generic future-narrative closers:** modal plus "become" plus "one of the most [adjective] [narrative / trend / chapter]". Grammatically a prediction, containing nothing testable. Replace with a falsifiable version or cut.
+- **Template phrases:** slot-fill constructions: "a [adjective] step towards [adjective] AI infrastructure", "Whether you're [X] or [Y]" (false breadth: pick the audience), "I recently had the pleasure of [verb]-ing".
+
+## Inflation and false depth
+
+- **Significance inflation:** "Marking a pivotal moment in the evolution of...". If the sentence still works after deleting the inflation clause, delete it.
+- **Aphorism formulas:** "X is the language of Y", "the architecture of trust". The shape does the persuading. Replace with the concrete claim. Carve-out: quotations and established idioms.
+- **"Real" and "actual" adjective inflation:** "Real on-chain tokenomics", "genuine utility". The intensifier implies the rest of the field is fake without saying what makes this one real. Carve-out: when the sentence names the contrast ("actual revenue from paying customers, not grants").
+- **Moral-adjective category errors:** moral adjectives glued to non-agentic nouns: "an honest shape", "flagged honestly". State the concrete property instead. Related: "the assumption stops being true" (assumptions break down, they do not flip), and gratuitous universal quantifiers ("taught in every first-year course").
+- **Novelty inflation:** treating established concepts as inventions: "he introduced a term", "a failure mode nobody's naming". Describe what the person did with the concept. Also flag invented labels, meaning pseudo-analytical compounds coined mid-sentence and never defined.
+- **Superficial -ing analyses:** "Symbolizing the region's commitment to progress, reflecting decades of investment". Also the declarative form: "this represents a broader shift". Show the consequence or cut.
+- **Promotional language:** tourism-brochure prose: "nestled within the breathtaking foothills", "a thriving ecosystem". Replace with plain description.
+- **Formulaic challenges:** "Despite challenges, X continues to thrive." Name the challenge and the response or cut the sentence.
+- **Emotional flatline:** "What surprised me most", "I was fascinated to discover", and the header form "Interesting part of the project:". If the thing is surprising, the content should show it.
+- **Lingering-attention claims:** "I can't stop thinking about this", "still thinking about this one". The claim is about the writer's attention and arrives before the reader has a reason to care. Carve-out: when the sentence says why the thing recurred.
+- **Self-labeling significance:** pointing back at an item to label it: "that last move is the contrarian one", "here's where it gets clever". If the move is contrarian, the description already showed it.
+- **Confidence calibration phrases:** "Interestingly", "Notably", "Importantly", "Undoubtedly". One in 2,000 words is fine; three in 500 is emphasis stacking. Related persuasive-authority tropes: "the real question is", "at its core", "fundamentally", "make no mistake".
+- **Filler phrases:** "It is important to note that", "In terms of", "The reality is that".
+
+## Manufactured credibility
+
+- **Vague attributions:** "Experts believe", "Studies show". Cite the source or state the claim directly.
+- **Vague third-party validation:** an unnamed authority plus a superlative: "independent testing confirms", "third-party benchmarks show we lead". Name the source, the test, and the result. Carve-out: a named benchmark, a linked report, a dated audit.
+- **Notability name-dropping:** piling on prestigious citations to borrow their weight. One specific reference beats four names. Related: historical analogy stacking ("like the printing press, the telegraph, and the internet before it").
+- **Speculative gap-filling:** guesses formatted as background: "is believed to have", "likely began his career in". Worse than a cutoff disclaimer because it hides the gap. Cut or source it.
+
+## Conversational-register tells
+
+- **Chatbot artifacts:** "I hope this helps!", "Certainly!", "Feel free to reach out", "In this article, we will explore...". Remove.
+- **Sycophantic tone:** "Great question!", "You're absolutely right!". Validation aimed at the reader rather than content.
+- **Narrated candor:** announcing disclosure instead of disclosing: "Two caveats I would rather flag than let you discover later:", "To be fully transparent:". Apply the deletion test: cut the frame and see whether any information is lost. Carve-outs: the substantive admission itself, and conventional conflict-of-interest disclosure that carries a material fact. Judgment-only, since every regex tight enough to spare the carve-outs stopped matching the tell.
+- **Acknowledgment loops:** "To answer your question", "The question of whether". Also opening a section by summarizing the previous one.
+- **Recap-flattery openers:** summarizing someone's own work back at them with praise before getting to the point. One plain clause of thanks, then substance.
+- **Wall-of-text replies:** in reply registers (issue comments, chat, DMs), roughly under 150 words with four or more sentences and no line break anywhere. Break at thought boundaries. Never fires on continuous long-form prose, where a dense paragraph is correct.
+- **Reasoning chain artifacts:** "Let me think step by step", "Breaking this down", "Here's my thought process". Scaffolding the reader does not need.
+- **Cutoff disclaimers:** "As of my last update", "I don't have access to real-time data". Never publish a sentence admitting the writer did not look something up.
+
+## Social-post tells
+
+- **Hashtag stuffing:** six or more hashtags on a short post, usually one specific tag mixed with broad category tags. Two or three specific tags maximum. Not counted: issue and PR references (`#88`), hex colors containing a digit, preprocessor directives, URL fragments, headings, and anything inside code.
+- **Social endorsement closers:** "This one is worth your time:", "Bookmark this.", "Thank me later." The endorsement could sit under any link. Say what the thing is and who it is for, then drop the call to action.
+- **Infomercial engagement hooks:** "The catch?", "Plot twist:", "Here's the thing." Mid-flow teasers that fake momentum. Delete the hook and state the thing. Same move in a fake-candid register: "Honestly?", "Real talk:" as standalone openers. Mid-sentence "honestly" is ordinary English.
+- **Bullet lists of bare noun phrases:** five or more consecutive items, each a short adjective-plus-noun phrase with no verb. The tell is the symmetry: every item the same shape, none of them checkable. Rewrite items as full claims. Does not apply to changelog entries, todo lists, or parameter docs.
+- **Manufactured punchlines:** three or more same-shape fragments in a row, each engineered to land like a closer. Keep the one that earns its emphasis and fold the rest into ordinary sentences.
+
+## Machine fingerprints
+
+These are deterministic publishing artifacts. A token proves a chat tool touched the text's pipeline, not who wrote the surrounding prose; treat each one as mandatory cleanup, never as an authorship verdict.
+
+- **Unfilled placeholders:** `[Your Name]`, `[INSERT SOURCE URL]`, `2025-XX-XX`, HTML comments with placeholder verbs. Treat as a publishing bug.
+- **Chat citation markup leaks:** `citeturn0search0`, `contentReference[oaicite:0]{index=0}`, `oai_citation`, `[attached_file:1]`, `grok_card`. Strip every token.
+- **AI-tool URL parameters:** `utm_source=chatgpt.com`, `utm_source=claude.ai`, `utm_source=perplexity.ai`, `referrer=grok.com`. Strip the AI-referrer tracking parameter from every URL that carries one, and leave the rest of the query string alone; a functional parameter (`?page=2`, `?v=4`) is not evidence of anything.
+
+## Rhythm and structure
+
+Structural regularity is the hardest signal to mask: a piece with every flagged word replaced and its rhythm untouched still reads as machine output. Weight rhythm above vocabulary when reviewing.
+
+- **Sentence-length uniformity:** most sentences landing in the 15 to 25 word band sounds robotic. Mix three-to-eight word sentences with 20-plus word ones. Fragments work.
+- **Paragraph-length uniformity:** vary deliberately. Some paragraphs should be one sentence.
+- **Missing first-person perspective:** where the genre carries a voice, relentless neutrality is itself a tell.
+- **Suspiciously clean grammar:** deliberate fragments, sentences opening with "And" or "But", and comma splices for effect belong to the author's voice. Keep them.
+- **Over-polishing:** editing out every irregularity pushes human writing toward the machine profile. Applying every rule at maximum strictness manufactures the uniformity the skill exists to remove.
+- **Read-aloud test:** text that a speech engine could read without sounding odd is probably too uniform.
+- **Vocabulary diversity:** in pieces over 200 words, type-token ratio (distinct types over total tokens) usually lands near 0.50 to 0.65 in English prose. Below 0.40 is worth a second look on general prose. It is not proof: narrow topics, reference material, and second-language writing all compress vocabulary legitimately. The fix is to broaden the subject matter, not to run a thesaurus over it.
+- **Diff-anchored writing:** docs that narrate a change instead of describing the thing: "this function was added to replace the previous approach". A reader without the commit history gets archaeology. Carve-out: changelogs, release notes, migration guides, and decision records are version-scoped by design.
+
+## Writer-side tests
+
+Judgment checks with no detectable surface form.
+
+- **Paragraph-reshuffle immunity:** can two body paragraphs swap without breaking the piece? If order does not matter, it is a list of points rather than an argument that builds. The fix is structural.
+- **Treadmill effect:** read each paragraph and ask what is new. If you could cut 40 to 60 percent and lose nothing, the prose is restating its premise in fresh words. Name the one fact or turn each paragraph contributes, or cut it.
+- **Rewrite from scratch instead of patching** when five or more vocabulary flags, three or more distinct pattern categories, and uniform sentence and paragraph length all appear together.

package/org-skills/avoid-ai-writing/references/profiles.md

@@ -0,0 +1,119 @@
+# Context and voice profiles
+
+Two independent axes. A context profile sets *how strict* to be for an audience. A
+voice profile sets *how the prose should sound*. Blunt for a blog and warm for docs
+are both valid combinations.
+
+## Context profiles
+
+- **`linkedin`**: short-form social. Punchy fragments and visual formatting matter.
+- **`blog`**: the default. Standard long-form prose, every rule at full strength.
+- **`technical-blog`**: long-form with code, architecture, and APIs. Technical terms get a pass.
+- **`investor-email`**: high-trust audience. Tighten everything. Promotional language is the biggest risk.
+- **`docs`**: documentation, READMEs, guides. Clarity over voice.
+- **`casual`**: chat messages, internal notes, quick replies. Only the worst offenders.
+
+### Auto-detection
+
+When no context is named, infer it:
+
+| Signal | Inferred context |
+|--------|-----------------|
+| Under 300 words plus hashtags or mentions | `linkedin` |
+| Code blocks, API references, or architecture | `technical-blog` |
+| Salutation plus investor or fundraising language | `investor-email` |
+| Step-by-step instructions, parameter docs, README structure | `docs` |
+| No strong signals | `blog`, the safest default |
+
+Say which profile you inferred and why. The writer can override.
+
+### Tolerance matrix
+
+Rules absent from the table apply at full strength everywhere. "Skip" means the rule
+does not apply to that register. "Extra strict" means flag borderline instances too:
+in an investor email, one "thriving ecosystem" can undermine the whole message.
+
+| Rule | linkedin | blog | technical-blog | investor-email | docs | casual |
+|------|----------|------|----------------|----------------|------|--------|
+| Em dashes | relaxed (2/post OK) | strict | strict | strict | relaxed | skip |
+| Bold overuse | relaxed (bold hooks OK) | strict | strict | strict | relaxed | skip |
+| Emoji in headers | relaxed (1-2 end-of-line OK) | strict | strict | strict | skip | skip |
+| Excessive bullets | skip (lists work on LinkedIn) | strict | relaxed (technical lists OK) | strict | skip (lists are docs) | skip |
+| Hedging | strict | strict | relaxed ("may" is accurate in technical) | strict | relaxed | skip |
+| Word table (full list) | strict | strict | **partial** (see below) | strict | relaxed | P0 only |
+| Promotional language | relaxed (some sell is expected) | strict | strict | **extra strict** | strict | skip |
+| Significance inflation | strict | strict | strict | **extra strict** | relaxed | skip |
+| Copula avoidance | skip | strict | relaxed | strict | skip | skip |
+| Uniform paragraph length | skip (short-form) | strict | strict | strict | relaxed | skip |
+| Numbered list inflation | relaxed | strict | relaxed | strict | skip | skip |
+| Rhetorical questions | relaxed (1 as hook OK) | strict | strict | strict | strict | skip |
+| Transition phrases | skip (short-form) | strict | strict | strict | relaxed | skip |
+| Generic conclusions | skip | strict | strict | **extra strict** | skip | skip |
+| Hashtag stuffing | strict | strict | strict | **extra strict** | skip (no hashtags in docs) | skip |
+| Bullet-NP lists | strict | strict | relaxed (technical option lists OK) | strict | relaxed (parameter lists OK) | skip |
+| Tier 3 phrase clustering | strict | strict | strict | **extra strict** | relaxed | skip |
+| Future-narrative closers | strict | strict | strict | **extra strict** | skip | skip |
+| Social endorsement closers | strict (the LinkedIn share-post tell) | strict | strict | strict | skip | relaxed (1 OK in a DM) |
+| Hedge-stacked predictions | strict | strict | relaxed ("could" is hedged accuracy) | **extra strict** | relaxed | skip |
+| Real/actual inflation | strict | strict | strict | **extra strict** | relaxed | skip |
+| Moral-adjective category errors | strict | strict | relaxed | strict | relaxed | skip |
+| Invented contrast-pair mirroring | strict | strict | relaxed | strict | relaxed | skip |
+| Subjectless fragments and agentless passives | relaxed (short-form fragments are the register) | strict | relaxed | strict | skip (fragment lists are docs) | skip |
+
+**Technical-blog word table exceptions.** These terms carry legitimate technical meaning
+and stay unflagged in technical context: `robust`, `comprehensive`, `seamless`, `ecosystem`,
+`leverage` (when the subject is actual platform leverage or APIs), `facilitate`, `underpin`,
+`streamline`. Still flag `delve`, `tapestry`, `beacon`, `embark`, `testament to`,
+`game-changer`, and `harness` there.
+
+One scoping rule lives in the pattern itself rather than in this table. Wall-of-text
+replies fire only in conversational reply registers, and a plain issue comment
+auto-detects as `blog`, so the judgment has to sit with the rule.
+
+## Voice profiles
+
+Voice is optional. With no profile named, infer the register from the input and avoid
+imposing a persona on text that already has one. Each profile is a set of concrete
+targets rather than a vibe. Targets shape what survives the edit; they never authorize
+adding stance, personality, or facts the source lacks. The rewrite guardrails in
+`SKILL.md` bind here too.
+
+**`casual`**: contractions throughout, since their absence reads stiff. Short sentences,
+averaging 14 words or fewer. Fragments allowed. Keep first-person and concrete
+touches where the source has them; never add one it lacks. Near-zero jargon. Keep warm hedges ("honestly", "I think") and cut corporate ones
+("it's worth noting"). Fits blog posts, social, community writing.
+
+**`professional`**: active voice for most sentences. Vary sentence length. Prefer a concrete
+claim (a number, a name, a date) over "experts say" when the source provides one. Keep
+the ask explicit where the source makes one; never invent facts or an ask.
+Low tolerance for hedging. Fits LinkedIn, investor email, pitches.
+
+**`technical`**: prefer plain copulatives ("X is Y") over inflated substitutes ("serves as",
+"stands as a testament to"). One idea per sentence, imperative mood for instructions.
+Jargon is fine when defined on first use. Tables and lists only where the content is
+genuinely list-shaped. Fits docs and technical blogs.
+
+**`warm`**: address the reader directly where the source already speaks to them, and
+keep its acknowledgment rather than adding one. Cut intensifiers
+("very", "truly", "incredibly") in favor of stronger verbs. No performative-empathy openers.
+Medium sentences of 15 to 20 words for an unhurried cadence. Fits mentorship, onboarding,
+thank-yous.
+
+**`blunt`**: lead with the claim and cut the windup. Dashes are rare here; use periods for
+emphasis. No padding to reach a rule of three. Near-zero hedging, and flag "may / could /
+potentially" stacks. Short declaratives with the occasional long sentence for contrast.
+Fits decision memos, thought leadership, hard feedback.
+
+**Calibrate to a sample instead.** Given a sample of the writer's own work, analyze its
+sentence-length pattern, contraction rate, paragraph openings, and recurring word choices,
+then match those rather than a named profile. Do not upgrade their vocabulary: if they
+write "stuff" and "things", keep that register.
+
+## How the axes compose
+
+A voice target always applies, even where a context profile would skip that category.
+Technical voice still prefers plain copulatives in a casual context that otherwise ignores
+copula avoidance. Where both axes govern the same rule and agree, they reinforce each other.
+Where they disagree, resolve toward the stricter of the two: a warm voice on `docs` still
+does not get decorative tables. Sensible default pairings are casual with casual,
+professional with linkedin or investor-email, technical with docs or technical-blog.

package/org-skills/avoid-ai-writing/references/word-tiers.md

@@ -0,0 +1,220 @@
+# Word tiers
+
+112 single-word and short-phrase entries across three tiers, plus 10 multi-word
+boilerplate phrases. The tier says how much weight a hit carries, not how hard the
+fix is. Counts here are derived from the upstream tables and enforced against them
+in the upstream CI.
+
+**Match inflected forms.** Each entry covers the listed word and its morphological
+variants: adverb, gerund, plural, comparative, and verb conjugations. `delve` covers
+`delving`, `leverage` covers `leveraging`. When a variant carries a separate honest
+sense (`real` meaning factual rather than the intensifier in "a real improvement"),
+judge by context instead of matching blindly.
+
+The tiering is adapted from [brandonwise/humanizer](https://github.com/brandonwise/humanizer)'s
+vocabulary research, which reduces false positives on words that are fine alone and
+suspicious in clusters.
+
+## Tier 1: always replace
+
+Tier 1 splits into two bands. Both get replaced and the edit is identical. What
+differs is what a hit *means*.
+
+**1A, AI frequency markers.** Words claimed to appear far more often in machine text
+than in human writing. A cluster of these is evidence about how a passage was produced.
+
+**1B, clarity edits.** Wordiness and inflated formality. Replacing them is good writing
+regardless of who wrote the sentence, and a 1B hit is **not** evidence of machine
+authorship. Measured against 257 paragraphs of verified pre-2023 human prose, 1B entries
+fire on ordinary professional writing at a meaningful rate: `in order to`, `utilize`,
+`commence`, `ascertain`, and `endeavor` are simply the words some people reach for.
+Weight them like Tier 2 and keep them out of any density signal, so a wordiness fix can
+never push a document toward an AI classification. In detect mode, report the two bands
+separately.
+
+Caveat worth keeping visible: the "appears far more often in AI text" claim behind 1A is
+inherited rather than measured. It traces to brandonwise/humanizer, which states a 5 to 20x
+ratio without publishing a method or dataset. Treat 1A as a well-supported convention until
+someone measures the ratios against a machine-written corpus.
+
+### Tier 1A: AI frequency markers (49 entries)
+
+| Replace | With |
+|---|---|
+| delve / delve into | explore, dig into, look at |
+| landscape (metaphor) | field, space, industry, world |
+| tapestry | (describe the actual complexity) |
+| realm | area, field, domain |
+| paradigm | model, approach, framework |
+| embark | start, begin |
+| beacon | (rewrite entirely) |
+| testament to | shows, proves, demonstrates |
+| robust | strong, reliable, solid |
+| comprehensive | thorough, complete, full |
+| cutting-edge | latest, newest, advanced |
+| leverage (verb) | use |
+| pivotal | important, key, critical |
+| underscores | highlights, shows |
+| meticulous / meticulously | careful, detailed, precise |
+| seamless / seamlessly | smooth, easy, without friction |
+| game-changer / game-changing | describe what specifically changed and why it matters |
+| hit differently / hits different | (say what specifically changed, or cut) |
+| watershed moment | turning point, shift (or describe what changed) |
+| marking a pivotal moment | (state what happened) |
+| the future looks bright | (cut — say something specific or nothing) |
+| only time will tell | (cut — say something specific or nothing) |
+| nestled | is located, sits, is in |
+| vibrant | (describe what makes it active, or cut) |
+| thriving | growing, active (or cite a number) |
+| despite challenges… continues to thrive | (name the challenge and the response, or cut) |
+| showcasing | showing, demonstrating (or cut the clause) |
+| deep dive / dive into | look at, examine, explore |
+| unpack / unpacking | explain, break down, walk through |
+| bustling | busy, active (or cite what makes it busy) |
+| intricate / intricacies | complex, detailed (or name the specific complexity) |
+| complexities | (name the actual complexities, or use "problems" / "details") |
+| ever-evolving | changing, growing (or describe how) |
+| enduring | lasting, long-running (or cite how long) |
+| daunting | hard, difficult, challenging |
+| holistic / holistically | complete, full, whole (or describe what's included) |
+| actionable | practical, useful, concrete |
+| impactful | effective, significant (or describe the impact) |
+| learnings | lessons, findings, takeaways |
+| thought leader / thought leadership | expert, authority (or describe their actual contribution) |
+| best practices | what works, proven methods, standard approach |
+| at its core | (cut — just state the thing) |
+| synergy / synergies | (describe the actual combined effect) |
+| interplay | relationship, connection, interaction |
+| keen (as intensifier) | interested, eager, enthusiastic (or cut — just state the interest) |
+| genuinely / genuine (as intensifier) | (cut — just state the fact) |
+| symphony (metaphor) | (describe the actual coordination or combination) |
+| embrace (metaphor) | adopt, accept, use, switch to |
+| load-bearing *(metaphor)* | essential, critical, necessary — or say what breaks if you remove it |
+
+**Hyphen required:** unhyphenated "load bearing" is ordinary English ("the load bearing
+down on the bridge"). Only the hyphenated compound is the tell.
+
+**Construction carve-out:** `load-bearing` before a literal structural noun (`wall`, `beam`,
+`column`, `joist`, `truss`, `footing`, `slab`, `stud`, `masonry`, `lintel`, `rafter`,
+`girder`, `capacity`), optionally with one material or position adjective in between, is
+standard building terminology. Abstract-capable nouns (`structure`, `element`, `frame`,
+`foundation`) stay flaggable, so "the load-bearing structure of his argument" still fires.
+
+### Tier 1B: clarity edits (10 entries)
+
+Wordiness and formality, not authorship evidence. Same fix, weaker claim.
+
+| Replace | With |
+|---|---|
+| utilize | use |
+| in order to | to |
+| due to the fact that | because |
+| serves as | is |
+| features (verb) | has, includes |
+| boasts | has |
+| presents (inflated) | is, shows, gives |
+| commence | start, begin |
+| ascertain | find out, determine, learn |
+| endeavor | effort, attempt, try |
+
+## Tier 2: flag when two or more appear in one paragraph (40 entries)
+
+These words are legitimate alone. Two or more together usually means the paragraph
+needs a rewrite.
+
+| Replace | With |
+|---|---|
+| harness | use, take advantage of |
+| navigate / navigating | work through, handle, deal with |
+| foster | encourage, support, build |
+| elevate | improve, raise, strengthen |
+| unleash | release, enable, unlock |
+| streamline | simplify, speed up |
+| empower | enable, let, allow |
+| bolster | support, strengthen, back up |
+| spearhead | lead, drive, run |
+| resonate / resonates with | connect with, appeal to, matter to |
+| revolutionize | change, transform, reshape (or describe what changed) |
+| facilitate / facilitates | enable, help, allow, run |
+| underpin | support, form the basis of |
+| nuanced | specific, subtle, detailed (or name the actual nuance) |
+| crucial | important, key, necessary |
+| multifaceted | (describe the actual facets, or cut) |
+| ecosystem (metaphor) | system, community, network, market |
+| myriad | many, numerous (or give a number) |
+| plethora | many, a lot of (or give a number) |
+| encompass | include, cover, span |
+| catalyze | start, trigger, accelerate |
+| reimagine | rethink, redesign, rebuild |
+| galvanize | motivate, rally, push |
+| augment | add to, expand, supplement |
+| cultivate | build, develop, grow |
+| illuminate | clarify, explain, show |
+| elucidate | explain, clarify, spell out |
+| juxtapose | compare, contrast, set side by side |
+| paradigm-shifting | (describe what actually shifted) |
+| transformative / transformation | (describe what changed and how) |
+| cornerstone | foundation, basis, key part |
+| paramount | most important, top priority |
+| poised (to) | ready, set, about to |
+| burgeoning | growing, emerging (or cite a number) |
+| nascent | new, early-stage, emerging |
+| quintessential | typical, classic, defining |
+| overarching | main, central, broad |
+| quietly | cut, or name the concrete contrast |
+| deeply *(significance collocations only — "deeply integrated," "deeply committed," "deeply rooted"; literal uses like "deeply nested" or "cares deeply" never count toward a cluster)* | cut, or name what specifically runs deep |
+| underpinning / underpinnings | basis, foundation, what supports |
+
+## Tier 3: flag only at high density (13 entries)
+
+Normal words. Flag them only when the text is saturated, which signals that vague
+praise filled the space where specifics belonged. Roughly 3 percent of total words is
+the threshold worth a second look.
+
+| Word | What to do |
+|---|---|
+| significant / significantly | Replace some with specifics: numbers, comparisons, examples |
+| innovative / innovation | Describe what's actually new |
+| effective / effectively | Say how or cite a metric |
+| dynamic / dynamics | Name the actual forces or changes |
+| scalable / scalability | Describe what scales and to what |
+| compelling | Say why it compels |
+| unprecedented | Name the precedent it breaks (or cut) |
+| exceptional / exceptionally | Cite what makes it an exception |
+| remarkable / remarkably | Say what's worth remarking on |
+| sophisticated | Describe the sophistication |
+| instrumental | Say what role it played |
+| world-class / state-of-the-art / best-in-class | Cite a benchmark or comparison |
+| verbatim | Usually redundant with the verb ("copies X verbatim" = "copies X") — cut it. If the exactness marks a contrast, name it: byte-for-byte, word for word, unchanged. Term of art in legal/research/QA registers ("verbatim transcript / record / testimony"), so weigh density in that context before flagging |
+
+## Tier 3 phrases: flag at density or in clusters (10 entries)
+
+Multi-word boilerplate that is individually unobjectionable and stacks heavily in
+generated content. Crypto, web3, DePIN, and AI infrastructure reviews are the worst
+offenders. Two rules apply: the same phrase used twice, and three or more *distinct*
+phrases from this table in one piece even when each appears once. The second rule
+catches the shape models take when they vary their own boilerplate to seem less
+repetitive.
+
+| Phrase | What to do |
+|---|---|
+| emerging sector / emerging space / emerging category | Name the actual sector or what's emerging about it |
+| the integration of (X with Y) | Describe what's being integrated and what changes for the user |
+| the intersection of (X and Y) | Pick the specific overlap that matters or cut the framing |
+| community-driven | Name what the community does. "Community-driven" alone is filler |
+| long-term sustainability | Cite the time horizon and the constraint. "Long-term" is hand-waving |
+| user engagement | Name the action. "Engagement" is a wrapper around clicks/comments/retention |
+| decentralized compute | Specify the architecture or cut. The phrase has become a category label, not a claim |
+| (sustainable) reward emissions | Cite the emission schedule and the sink |
+| tokenized incentive structures | Describe the actual mechanism (vesting, gauge, bonded LP, etc.) |
+| designed for long-term [X] | Cut "designed for" — either it is or it isn't. Then state the property |
+
+## Context adjustments
+
+`technical-blog` gives legitimate technical meaning back to `robust`, `comprehensive`,
+`seamless`, `ecosystem`, `leverage` (actual platform leverage or APIs), `facilitate`,
+`underpin`, and `streamline`. Still flag `delve`, `tapestry`, `beacon`, `embark`,
+`testament to`, `game-changer`, and `harness` there.
+
+`docs` relaxes the full table. `casual` drops to P0 patterns only. See
+`references/profiles.md` for the whole matrix.

package/{dist/src/orgrt/role-skills/backend-dev.md → org-skills/backend-dev/SKILL.md}

@@ -1,3 +1,11 @@
+---
+name: backend-dev
+description: "Backend Dev role best practices. Builds and maintains server-side services, APIs, and database logic that are correct, secure, and performant under real load."
+tags: ["engineering"]
+tools: ["monograph_query","monograph_context","monograph_impact"]
+license: Apache-2.0
+source: https://github.com/monoes/monomind
+---
 # Backend Dev — Best Practices
 
 ## Focus

package/org-skills/backend-patterns/LICENSE.txt

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Affaan Mustafa
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.