# Yun Write Skill > Use when authoring or editing a skill in this workspace — deciding how it is invoked, what lives in SKILL.md versus a linked file, naming it, or diagnosing why a skill misfires, bloats, or fires unreliably. - Skill: `matiinchauspe/yun-write-skill` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add matiinchauspe/yun-write-skill` - Raw SKILL.md: https://api.skillmd.com/api/skills/matiinchauspe/yun-write-skill/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Coding & Dev Tools - Author: matiinchauspe (https://skillmd.com/u/matiinchauspe) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/matiinchauspe/yun-write-skill --- # yun-write-skill A skill exists to wrangle determinism out of a stochastic system. **Predictability** — the agent taking the same _process_ every run, not producing the same output — is the root virtue; every lever below serves it. **Bold terms** are defined in [`GLOSSARY.md`](GLOSSARY.md); look them up there for the full meaning. ## In this workspace This is a harness skill. Two workspace rules bind every skill you author here — do not restate them, read them at their source: - **Naming:** follow the `yun--` convention in `AGENTS.md`. The directory name must match the skill `name`. - **Home:** harness skills live in `skills/`; they are always active when your agent opens from the workspace root. A repo's own skills are secondary. **Before authoring or diagnosing, `recall` the harness's `friction` topic** (through `skills/_shared/memory-convention.md`) — sessions working *other* projects file there what the harness got wrong. It is the harness's only supply of measured pain. The rest of this skill is the quality bar those skills — and the `skills/_shared/` contracts they lean on — are judged against; a contract is authored, disclosed, pruned and traversed by the same rules. ## Invocation Two choices, trading different costs: - A **model-invoked** skill keeps a **description**, so the agent can fire it autonomously _and_ other skills can reach it (you can still type its name too). It contributes to **context load** — the description sits in the window every turn. Mechanics: omit `disable-model-invocation`, and write a model-facing description with rich trigger phrasing ("Use when the user wants…, mentions…"). - A **user-invoked** skill strips the description from the agent's reach: only you, typing its name, can invoke it — and no other skill can. Zero context load, but it spends **cognitive load**: _you_ are the index that must remember it exists. Mechanics: set `disable-model-invocation: true`; the `description` becomes human-facing — a one-line summary, trigger lists stripped. Pick model-invocation only when the agent must reach the skill on its own, or another skill must. If it only ever fires by hand, make it user-invoked and pay no context load. When user-invoked skills multiply past what you can remember, that piled-up cognitive load is cured by a **router skill**: one user-invoked skill that names the others and when to reach for each. ## Writing the description A model-invoked **description** does two jobs — state what the skill is, and list the **branches** that should trigger it. Every word increases **context load**, so a description earns even harder pruning than the body: - **Front-load the skill's leading word** — the description is where it does its invocation work. - **One trigger per branch.** Synonyms that rename a single branch are **duplication** — "build features using TDD … asks for test-first development" is one branch written twice. Collapse them; keep only genuinely distinct branches. - **Cut identity that's already in the body.** Keep the description to triggers, plus any "when another skill needs…" reach clause. ## Information hierarchy A skill is built from two content types — **steps** and **reference** — that mix freely: a skill can be all steps, all reference, or both. The core decision is which to use and where each sits on the **information hierarchy**, a ladder ranked by how immediately the agent needs the material: 1. **In-skill step** — an ordered action in `SKILL.md`, the primary tier: what the agent does, in order. Each step ends on a **completion criterion**, the condition that tells the agent the work is done. Make it _checkable_ (can the agent tell done from not-done?) and, where it matters, _exhaustive_ ("every modified model accounted for", not "produce a change list") — a vague criterion invites **premature completion**. 2. **In-skill reference** — a definition, rule, or fact in `SKILL.md`, consulted on demand. Often a legitimately flat peer-set (every rule of a review on one rung) — a fine arrangement, not a smell. _This skill is all reference._ 3. **External reference** — reference pushed out of `SKILL.md` into a separate file, reached by a **context pointer**, loaded only when the pointer fires. (Spans _disclosed_ reference — a sibling file like `GLOSSARY.md`, still part of the skill — through fully **external reference** that lives outside the skill system and any skill can point at.) A demanding completion criterion drives thorough **legwork** — the digging the agent does within the work — whether the skill has steps or not, since "every rule applied" binds flat reference just as "every step done" binds a sequence. Push too little down and the top bloats; push too much and you hide material the agent actually needs. That tension is the whole decision. **Progressive disclosure** is the move down the ladder — out of `SKILL.md` into a linked file — so the top stays legible. Mechanics: a linked `.md` file in the skill folder, named for what it holds (this skill discloses its full definitions to `GLOSSARY.md`). **When to pull the lever:** the length itself is the trigger — treat a single reference block past ~50 lines, or a `SKILL.md` past ~150–200 lines (~500 words for a frequently-loaded skill), as **sprawl** and disclose it; no other justification is needed. Some skills are used in more than one way, and each distinct way is a **branch** — different runs taking different paths through the skill. Branching is the cleanest disclosure test: inline what every branch needs, and push behind a pointer what only some branches reach. A **context pointer**'s _wording_, not its target, decides when and how reliably the agent reaches the material. Disclosing moves antecedents behind that pointer and leaves every branch but the one that opens the file holding them, so close a disclosure with a **traversal**. Where the ladder decides _how far down_ a piece sits, **co-location** decides _what sits beside it_ once there: keep a concept's definition, rules, and caveats under one heading rather than scattered, so reading one part brings its neighbours with it. ## When to split **Granularity** is how finely you divide skills, and each cut spends one of the two loads, so split only when the cut earns it. Two cuts: - **By invocation** — split off a **model-invoked** skill when you have a distinct **leading word** that should trigger it on its own, or another skill must reach it. You pay **context load** for the new always-loaded **description**, so that independent reach has to be worth it. - **By sequence** — split a run of **steps** when the steps still ahead (a step's **post-completion steps**) tempt the agent to rush the one in front of it (**premature completion**). Keeping them out of view encourages the agent to do more **legwork** on the current task. ## Pruning Keep each meaning in a **single source of truth**: one authoritative place, so changing the behaviour is a one-place edit. Check every line for **relevance**: does it still bear on what the skill does? Then hunt **no-ops** sentence by sentence, not just line by line: run the no-op test on each sentence in isolation, and when one fails, delete the whole sentence rather than trim words from it. Be aggressive — most prose that fails should go, not be rewritten. ## Leading words A **leading word** is a compact concept already living in the model's pretraining that the agent thinks with while running the skill (e.g. _lesson_, _fog of war_, _tracer bullets_). Repeated throughout the text (though not necessarily — a strong leading word might only be needed once), it accumulates a distributed definition and anchors a whole region of behaviour in the fewest tokens, by recruiting priors the model already holds. It serves predictability twice. In the body it anchors _execution_: the agent reaches for the same behaviour every time the word appears. In the description it anchors _invocation_: when the same word lives in your prompts, docs, and code, the agent links that shared language to the skill and fires it more reliably. Hunt for opportunities to refactor skills to use leading words. A triad spelled out at three sites (**duplication**), a description spending a sentence to gesture at one idea — each is a passage begging to **collapse** into a single token. Examples include: - "fast, deterministic, low-overhead" -> _tight_ — one quality restated across a phase — into a single pretrained word (a _tight_ loop). - "a loop you believe in" -> _red_ — converts a fuzzy gate into a binary observable state (the loop goes _red_ on the bug, or it doesn't). You win twice over: fewer tokens, _and_ a sharper hook for the agent to hang its thinking on. Assume every skill is carrying restatements that leading words retire — go find them. ## Failure modes Use these to diagnose issues the user may be having with the skill. - **Premature completion** — ending a step before it's genuinely done, attention slipping to _being done_. Defence, in order: sharpen the completion criterion first (cheap, local); only if it is irreducibly fuzzy _and_ you observe the rush, hide the post-completion steps by splitting (the sequence cut). - **Duplication** — the same meaning in more than one place. Costs maintenance and tokens, and inflates a meaning's prominence on the ladder past its real rank. - **Sediment** — stale layers that settle because adding feels safe and removing feels risky. The default fate of any skill without a pruning discipline. - **Sprawl** — a skill simply too long, even when every line is live and unique. Hurts readability and maintainability and wastes tokens. The cure is the ladder: disclose **reference** behind pointers, and split by **branch** or sequence so each path carries only what it needs. - **No-op** — a line the model already obeys by default, so you pay load to say nothing. The test: does it change behaviour versus the default? A weak leading word (_be thorough_ when the agent is already thorough-ish) is a no-op; the fix is a stronger word (_relentless_), not a different technique. - **Negation** — steering by prohibition backfires: _don't think of an elephant_ names the elephant and makes it more available, not less. Prompt the **positive** — state the target behaviour so the banned one is never spoken; keep a prohibition only as a hard guardrail you can't phrase positively, and even then pair it with what to do instead. - **Dead branch** — a **branch** that cannot be completed as written: it closes on a **completion criterion** only a different branch can satisfy, or it reads a reference that resolves nowhere along this path. The defect lives on a path, not in a passage — every passage is correct read alone, so re-reading the edited region cannot surface it, and neither can a reviewer of the version before the edit. A skill is a program with no test suite, and reading one is not running it. Defence: a **traversal**. The trigger is the skill's _shape_, not your edit: traverse every branch of any skill you touch that carries more than one, including the branches no edit went near — a defect born in an earlier edit sits dormant on an untravelled path until someone walks it. A linear skill has no path to walk, so traversing it is a **no-op**. ## Attribution Adapted for this workspace from **Matt Pocock's `writing-great-skills`** (github.com/mattpocock/skills, MIT). Changes: renamed to the `yun--` harness convention, switched to model-invocation so it applies automatically while authoring, workspace-mechanics deferred to `AGENTS.md`, and the OpenAI agent wiring and aihero.dev/`ask-matt` ecosystem references dropped.