tag-by-topic
$
npx mdskill add lyndonkl/claude/tag-by-topicAssigns 1-4 topic tags using keyword and title matching.
- Prevents tag sprawl by requiring existing matches or logged additions.
- Depends on shared-context/topic-ledger.md for controlled vocabulary.
- Scores tags via keyword frequency and weighted title presence.
- Returns selected topics and proposed new tags in JSON format.
SKILL.md
.github/skills/tag-by-topicView on GitHub ↗
---
name: tag-by-topic
description: Assigns 1-4 topic tags to a seed body from the controlled vocabulary in substacker shared-context/topic-ledger.md. Prevents tag sprawl at small-corpus scale by requiring existing-tag match or logged addition. Uses keyword + title match; logs near-miss candidates to pending-tags. Use after format normalization and before dedupe. Trigger keywords: tag, topics, categorize, classify, taxonomy, controlled vocabulary, topic ledger.
---
# Tag by Topic
## Table of Contents
- [Workflow](#workflow)
- [Controlled vocabulary rules](#controlled-vocabulary-rules)
- [Worked example](#worked-example)
- [Guardrails](#guardrails)
**Related skills:** Called by `ingest-inbox-item` step 2. Writes to `topic-ledger.md` pending-tags section when proposing additions. Upstream of `update-topic-ledger` (which records the finalized counts per tag).
## Workflow
```
Tag one seed:
- [ ] Step 1: Read shared-context/topic-ledger.md — collect current tag set
- [ ] Step 2: Score each existing tag against body + title (keyword + weighted title match)
- [ ] Step 3: Select top 1-4 tags above threshold
- [ ] Step 4: If top-1 score below threshold, propose new tag in kebab-case; log to pending-tags
- [ ] Step 5: Return {topics: [str], pending_tags: [str]}
```
### Step 2: Scoring
For each existing tag, compute:
- `keyword_match` = number of times tag-words appear in body (normalized by body length)
- `title_match` = 3× if tag-words appear in title, else 0
- `score = keyword_match + title_match`
Threshold for selection: `score ≥ 0.3` (tunable — start at 0.3, adjust if tagging is noisy).
### Step 3: Select top 1–4
- If 0 tags above threshold → assign `unclassified` and log this as a signal that vocabulary may need extension.
- If 1–4 tags above threshold → return them, highest-score first.
- If >4 above threshold → keep top 4 by score.
### Step 4: New-tag proposal
If the top-1 score is below threshold, the body is about something the vocabulary doesn't cover. Propose a new tag:
- Extract 2–4 candidate keywords from title + first paragraph.
- Kebab-case: lowercase, hyphen-separated, `^[a-z0-9]+(-[a-z0-9]+)*$`.
- Append to `topic-ledger.md#pending-tags`:
```
- tag: {proposed-tag}
proposed_at: YYYY-MM-DD
seed_id: {seed-id}
justification: "{one-line reason}"
```
- Still return the closest existing tag for the seed's `topics`, so the seed is not untagged.
Never silently add to canonical vocabulary. The writer reviews pending-tags and promotes.
## Controlled vocabulary rules
- Max 4 tags per seed (enforced).
- Min 1 tag (enforced; use `unclassified` if truly nothing fits).
- Tag names: `^[a-z0-9]+(-[a-z0-9]+)*$`.
- New-tag additions require logged pending entry. Never silent.
- Tags are internal taxonomy — NOT Substack tags (which are a separate platform feature).
## Worked example
**Input seed body** (dropout-as-ensemble):
> had a thought while running — dropout is secretly an ensemble method. each forward pass is a different sub-network. so at test time when you turn dropout off and scale, you're averaging predictions across exponentially many thinned networks. this is why it generalizes. not "regularization" in the L2 sense. more like bagging.
**Existing tags** (from topic-ledger.md): `regularization`, `attention-mechanism`, `rag`, `emergence`, `kalshi`, `ipl-cricket`, `pathology-ai`.
**Scoring**:
- `regularization`: keyword "regularization" in body → 0.4. Score: 0.4.
- `ensembling`: NOT in vocabulary → propose.
- `dropout`: NOT in vocabulary → propose.
**Output**:
- `topics: [regularization]` (only one existing tag scored above threshold)
- `pending_tags: [ensembling, dropout]` logged to topic-ledger with this seed-id.
The writer reviews pending and either promotes both to canonical or renames to existing synonyms.
## Guardrails
1. Maximum 4 tags — hard enforced.
2. Minimum 1 tag — `unclassified` if nothing fits.
3. New tag proposal is always logged; never silent addition.
4. Tag names must match `^[a-z0-9]+(-[a-z0-9]+)*$`. Reject any other form.
5. Do not remove existing tags from the ledger. If a tag becomes cold, that's a state change, not a removal.
6. Do not use `score-intuition-density` or `dedupe-against-corpus` logic here — this is topic-only.
## Quick reference
- Input: seed body + partial frontmatter.
- Output: `{topics: [str], pending_tags: [str]}`.
- Side effect: may append to `topic-ledger.md#pending-tags`.
More from lyndonkl/claude
- abstraction-concrete-examplesBuilds structured abstraction ladders that translate high-level principles into concrete, actionable examples across 3-5 levels. Bridges communication gaps, reveals hidden assumptions, and tests whether abstract ideas work in practice. Use when explaining concepts at different expertise levels, moving between abstract principles and concrete implementation, identifying edge cases by testing ideas against scenarios, designing layered documentation, decomposing complex problems into actionable steps, or bridging strategy-execution gaps.
- academic-letter-architectGuides the creation of evidence-based academic recommendation letters, reference letters, and award nominations that combine concrete examples, meaningful comparisons, and genuine enthusiasm. Use when writing recommendation letters for students, postdocs, or colleagues, or when user mentions recommendation letter, reference, nomination, letter of support, endorsement, or needs help with strong advocacy and comparative statements.
- adr-architectureDocuments significant architectural and technical decisions with full context, alternatives considered, trade-offs analyzed, and consequences understood. Creates a decision trail that helps teams understand why decisions were made. Use when choosing between technology options, making infrastructure decisions, establishing standards, migrating systems, or when user mentions ADR, architecture decision, technical decision record, or decision documentation.
- adverse-selection-priorProduces a Bayesian prior probability that an offered transaction is +EV for the recipient, given that the counterparty chose to propose it. Applies Akerlof market-for-lemons logic -- if they offered it, they believe it is +EV for them, so the prior that it is +EV for us is materially below 50%. Reusable across trade evaluation, waiver drops (another team dropping a player is also adverse selection), job-offer analysis, M&A, and any "someone offered me this" situation. Use when you receive an unsolicited trade/offer/proposal, analyzing incoming trade prior, evaluating why a counterparty proposed a deal, or when user mentions adverse selection, market for lemons, why did they offer this, incoming trade prior, they proposed it, Bayesian adjustment on received offer.
- alignment-values-north-starCreates actionable alignment frameworks that give teams a shared North Star (direction), values (guardrails), and decision tenets (behavioral standards). Enables autonomous decision-making while maintaining organizational coherence. Use when starting new teams, scaling organizations, defining culture, establishing product vision, resolving misalignment, creating strategic clarity, or when user mentions North Star, team values, mission, principles, guardrails, decision framework, or cultural alignment.
- analogy-weight-checkFor every analogy in a substacker draft, verifies it carries mechanical weight — the analogy does real work explaining the mechanism, not merely decorates it. Cross-references analogy-catalog.md for novelty (is this analogy reused from a prior post?) and domain fit (biology > organizational > sports preferred; physics/military disfavored). Use whenever an analogy appears in the draft. Trigger keywords: analogy weight, decorative, mechanical weight, reused analogy, catalog check, metaphor check.
- answer-uncomfortable-questionTakes one strategic question about substacker ("should we launch paid?", "is this section dead?", "are we writing for the wrong audience?") and produces the mandatory evidence + reasoning + downside triad plus a recommendation. Used 3 times per Growth Strategist review. Trigger keywords: uncomfortable question, strategic question, evidence reasoning downside, triad.
- attribute-performanceFor each substacker post that materially over- or under-performs the rolling baseline (|z| ≥ 1.0), produces a plain-English attribution paragraph with calibrated confidence (high / medium / low / unexplained). Considers subject-line effect, topic zeitgeist, external share, day-of-week, length effect, and audience-notes signals. Labels unexplained outliers explicitly rather than fabricating a story. Use after compute-baseline when outlier posts exist. Trigger keywords: attribution, why did this post work, outlier explanation, performance analysis.
- auction-first-price-shadingComputes the optimal shaded bid for a first-price sealed-bid auction given a true private value, an estimate of the number of competing bidders N, and a value-distribution assumption. Implements the `(N-1)/N` equilibrium shading rule for uniform private values, adjusts for log-normal or empirical value distributions, layers a risk-aversion adjustment, and caps output against the bidder's remaining budget. Domain-neutral auction theory reusable across fantasy sports (baseball FAAB, NBA/NHL waiver auctions), prediction-market limit sizing, sealed procurement bids, and any blind-bid context. Use when user mentions "first-price auction bid", "sealed bid shading", "(N-1)/N", "FAAB bid amount", "auction shading", "optimal bid first-price", "bid for sealed-bid", "blind bid sizing", or when downstream logic needs a principled shade factor rather than an ad-hoc heuristic.
- auction-winners-curse-haircutApplies a Bayesian haircut to a bid valuation for common-value auctions where winning is itself evidence the bidder over-estimated. Takes a raw valuation, a value-type classification (common_value / private_value / mixed), the number of informed bidders N, and a signal-dispersion estimate, and returns an adjusted valuation. Domain-neutral and reusable across fantasy FAAB, prediction markets, M&A bids, ad-auction budgets, and any generic bidding context. Use when user mentions "winner's curse", "common value auction", "valuation haircut", "adverse valuation", "Bayesian bid adjustment", or "over-paying in auction".