@victortomaili/skill-cli 0.6.6 → 0.6.8

package/CHANGELOG.md

@@ -11,6 +11,26 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Cursor adapter (`.cursor/rules` format) for `init -g` bootstrap.
 - Per-agent hook adapters for automatic `/X` triggering (push model).
 
+## [0.6.8] - 2026-07-08
+
+### Changed
+- AGENTS.md + `skill active`: reframed cost/style/speed skills as USER TRADE-OFFS
+  the agent must never pre-judge. The agent kept skipping proposals with "not
+  relevant to this task"; the new wording forbids deciding a trade-off for the
+  user ("not relevant here" = stealing the user's decision). Made the `skill active`
+  footer mechanical (LOAD/PROPOSE/SKIP grid) with an explicit callout: "cost/style
+  but not relevant here → that is the bug. Ask instead."
+
+## [0.6.7] - 2026-07-08
+
+### Changed
+- AGENTS.md START GATE: heavily compressed (605 → ~250 words) and made the
+  classification mechanical. cost/style/speed-affecting skills MUST be proposed on
+  EVERY task (those axes apply by definition); the rationalizations "not needed for
+  this task", "the user didn't request it", and "skip unless needed" are now
+  explicitly FORBIDDEN. Also de-hardcoded the caveman example (it overfit the test)
+  — the rule is now fully generic.
+
 ## [0.6.6] - 2026-07-08
 
 ### Changed

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@victortomaili/skill-cli",
-  "version": "0.6.6",
+  "version": "0.6.8",
   "description": "Cross-agent skill manager — one global store, clean agent folders, activation via skill.config",
   "license": "MIT",
   "author": "Victor Tomaili <victor@tomaili.com>",

package/src/commands/defaults.js

@@ -16,7 +16,7 @@ export function cmdActive() {
   const eff = computeEffective(installed, globalCfg, projCfg)
   const defs = new Set((globalCfg.defaults || []).map(d => String(d).toLowerCase()))
 
-  console.log(c.bold('skill active') + c.gray(' — active skills (descriptions only). AUTO-LOAD what is necessary or quality-improving; PROPOSE style/cost/shortcut skills and wait for confirmation.'))
+  console.log(c.bold('skill active') + c.gray(' — active skills (descriptions only). LOAD correctness/quality ones; PROPOSE every cost/style/speed one (a trade-off is the USER') + c.gray("'s call — never skip it)."))
   console.log()
 
   if (eff.length === 0) {
@@ -40,7 +40,13 @@ export function cmdActive() {
     if (s.description) console.log(c.gray('      ' + String(s.description).replace(/[\r\n]+/g, ' ').trim()))
   }
   console.log()
-  console.log(c.bold('→ Decide each skill by the criteria:') + c.gray(' affects ') + c.cyan('CORRECTNESS/QUALITY') + c.gray(' → load (') + c.cyan('skill cat <name>') + c.gray('); affects ') + c.cyan('COST/SPEED') + c.gray(' → propose; affects none → skip. Task type is irrelevant; when unsure → propose.'))
+  console.log(c.bold('→ For EACH skill above, decide in your reply:'))
+  console.log(c.gray('    LOAD    moves CORRECTNESS/QUALITY → run ') + c.cyan('skill cat <name>') + c.gray(' now.'))
+  console.log(c.gray('    PROPOSE moves COST, STYLE, or SPEED  → you MUST ASK "enable <name>?". A trade-off is the'))
+  console.log(c.gray('            USER') + c.gray("'s decision — these move an axis on EVERY reply, so NEVER skip one and"))
+  console.log(c.gray('            NEVER say "not relevant to this task" (that = you deciding for the user).'))
+  console.log(c.gray('    SKIP    moves NONE of the above.'))
+  console.log(c.yellow('  ⚠ If you wrote "cost/style but not relevant here" → that is the bug. Ask instead.'))
 }
 
 // `skill default <name>` — mark a skill as a default (active by default in every

package/src/lib/agents-md.js

@@ -24,42 +24,28 @@ Usage:
 - \`skill default <name>\` — mark a skill as a global default (active + auto-load)
 
 START GATE (mandatory): On the FIRST user message of a session, your VERY FIRST
-action — BEFORE ANYTHING ELSE: before you start thinking/reasoning about the task
-and before any tool call — MUST be to run \`skill active\`. It prints the skill
-CATALOG — every ACTIVE skill's name + FULL description (NOT the skill body).
-Read it, then for EACH skill judge by ONE question — does it affect the work's
-CORRECTNESS, QUALITY, COST (tokens), or SPEED? Decide from the description:
-- Affects CORRECTNESS or QUALITY → AUTO-LOAD it (\`skill cat <name>\`, silently).
-  These are how the job should be done — not optional.
-- Affects COST or SPEED only → PROPOSE it: tell the user what it does and apply
-  ONLY after they confirm (\`/X\` = confirm). These are trade-offs the user opts into.
-- Affects NONE of the four → skip it.
-ENFORCEMENT — state each decision OUT LOUD (load / propose / skip). Task type does
-NOT filter candidates: a COST/STYLE skill (e.g. a compression mode) affects cost on
-EVERY task — review, debugging, refactor, anything — so it is a candidate
-everywhere; "the user didn't ask" is NOT a reason to skip one that applies, and
-"not needed for this technical task" is wrong (style/cost are task-independent).
-Do not wait to be asked — propose proactively. When unsure whether a skill applies
-→ PROPOSE.
-PRIORITIES — when trading off how to do the work, rank: CORRECTNESS first, then
-QUALITY, then COST (cheap / token-efficient), then SPEED (fast). Never sacrifice
-correctness or quality for speed or cost.
-LOADED ≠ LISTED. A skill is loaded ONLY if you have run \`skill cat <name>\` for
-it THIS session. Seeing a skill in \`skill active\` / \`skill list\` does NOT load
-it — you then know only its name + description, nothing more. Never infer a skill
-is in your context from its name, the \u2605 marker, or an \`active\` status: if you
-haven't catted it, treat it as NOT loaded (then decide: cat it, or propose it).
-Do not assume a skill is already applied — \`active\`/\u2605-default = AVAILABLE, not
-APPLIED. ANY skill can be context-altering; judge from the description — there is
-no fixed list and no flag. Do this even if the task looks trivial: skills shape
-how you approach it, so you cannot reason about it correctly until you've reviewed
-them.
-
-Discovery (every message): On EVERY later user message, re-check the catalog
-against the new request; load any newly-relevant functional skill
-(\`skill cat <name>\`), and PROPOSE any newly-relevant context-altering one.
-Load each skill only ONCE per session (\`skill trigger <keyword>\` resolves a
-keyword to a skill).
+action — BEFORE ANYTHING ELSE (before thinking, before any tool call) — MUST be to
+run \`skill active\`. It lists each ACTIVE skill's name + FULL description (never the
+body). Read it, then for EACH skill classify which axis it moves and ACT — there is
+no discretion and no third option:
+- moves CORRECTNESS or QUALITY → load it now: \`skill cat <name>\`.
+- moves COST, SPEED, or response STYLE → PROPOSE it. These are TRADE-OFFS — the
+  USER decides them, NEVER you. Even if you are SURE one would hurt THIS task, you
+  still ASK: "Enable <name>? It <one-line benefit>." (apply only on yes). They move
+  an axis on EVERY task by definition (every reply has a token cost, a length, a
+  style, a speed), so the thoughts "not relevant to this task", "not needed here",
+  "the user didn't request it", and "skip unless needed" are FORBIDDEN — each one
+  is you STEALING the user's decision. Deciding a trade-off is not your role;
+  surfacing it is. A line in a table is NOT proposing — you must actually ask.
+- moves none of the above → skip.
+When unsure if a skill moves an axis → PROPOSE (ask). LOADED \u2260 LISTED: a skill is
+loaded only if you \`cat\`-ed it this session — listing it, its \u2605, or its \`active\`
+status is NOT loading; active/\u2605 means AVAILABLE, not APPLIED.
+PRIORITIES: correctness > quality > cost (cheap) > speed — never trade correctness
+or quality for speed or cost.
+Discovery: on EVERY later message, re-run this classification; load newly-relevant
+correctness/quality skills, PROPOSE newly-relevant cost/style/speed ones. Load each
+skill only ONCE per session (\`skill trigger <keyword>\` resolves a keyword).
 
 Triggers: when the user types \`/X\`, run \`skill trigger X\`.
 - Single match → apply the output directly.