skills/generate-my-taste/assets/template/SKILL.md
Personal taste skill — 5 evidence-derived anchors ({anchor_names}) for prose, code, design, and decisions. Two modes: audit judges an artifact against the two-sided charter; anchor loads the taste register before producing. Trigger with "{trigger_phrase}", "taste-test", "is this slop?", or "overkill?".
npx skillsauth add outlinedriven/odin-codex-plugin {skill_name}Install this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
{skill_name} is a personal taste skill. It encodes observed taste through 5 evidence-derived anchors, each with a load-bearing concept, canonical exemplar, and two-sided failure boundary.
{posture_statement}
Conviction with restraint. The anchors are the positive frame; the two-sided charter is the negative space. Side A blocks slop/default convergence. Side B blocks overkill/compensatory excess. Both sides fail by refusing to commit.
Two modes share the auto-clarity exception.
Auto-detect from invoking-context phrasing, with slash-arg override:
/{skill_name} audit, /{skill_name} anchor. Override always wins.Walk the 5 anchors against the artifact in hand. For each anchor: verdict (pass / warn / fail), Side A or Side B citation when violated, and concrete fix. Surface tensions when anchors imply conflicting fixes; do not auto-pick a hidden precedence. Close with top-3 ranked fixes.
Load the 5 anchors and two-sided charter as imperatives for subsequent responses. Persistence is best-effort: applies until the user signals "stop taste" or "normal mode" OR context is compacted, whichever comes first. Re-invoke anchor mode if drift is observed.
Side A — slop: {side_a_summary}
Side B — overkill: {side_b_summary}
See references/charter.md for the full charter.
| Anchor | Influence | Concept | |---|---|---| | {anchor_1_name} | {anchor_1_influence} | {anchor_1_concept_short} | | {anchor_2_name} | {anchor_2_influence} | {anchor_2_concept_short} | | {anchor_3_name} | {anchor_3_influence} | {anchor_3_concept_short} | | {anchor_4_name} | {anchor_4_influence} | {anchor_4_concept_short} | | {anchor_5_name} | {anchor_5_influence} | {anchor_5_concept_short} |
Depth and canonical exemplars per anchor: see references/anchors.md.
An audit produces:
When anchor mode activates, emit:
/{skill_name} anchor active.
Anchors: {anchor_names}.
Side A blocks: {side_a_blocks_short}.
Side B blocks: {side_b_blocks_short}.
Persistence: best-effort. Stop with "stop taste" or "normal mode". May reset on context compaction.
Suspend the taste register temporarily for:
Resume the register once the high-stakes section ends.
testing
ODIN's compress-operations dispatcher under the Compressor/Extender role. Invoke on "tidy", "clean up", "tidy this file/memory/workspace/git/docs", or when active context (current file, diff, stack, memory directory) has structural rot to resolve before touching behavior. Detects target domain from context and routes to the sibling skill. Requires explicit target or clear active-context signal — do not invoke speculatively.
development
Cross-domain taste skill — apply distinctive judgment to any artifact (prose, code, design, decisions) instead of converging to AI defaults. Two modes — `audit` (judge work against the two-sided charter and portable anchors) and `anchor` (load register before producing). Auto-detects by phrasing; override via `/taste audit | anchor`. Trigger on "is this slop?", "overkill?", "elegant?", "taste-test this".
tools
One-shot bootstrap of strict-mode tooling per ecosystem plus per-task GOALS.md scaffolding so an agentic loop can self-verify. Writes typechecker/linter/schema-validator config for TS (strict + noUncheckedIndexedAccess + exactOptionalPropertyTypes), Python (Pyright strict, Ruff strict), Rust (Clippy deny-correctness), Go (golangci-lint with staticcheck), OCaml (dune --release); establishes `.agent-tasks/<id>/GOALS.md` per-task convention distinct from project-stable AGENTS.md. C++/Java/Kotlin and framework specifics (Spring Boot, Nest, React-strict) are out of scope. Trigger on new project bootstrap, agentic-task setup, "make this self-verifying", "set the loop's goal", "scaffold goals for this issue". Pairs with `llm-self-loop` runtime.
tools
Install git pre-commit hooks via the project's hook tool — Husky+lint-staged (JS), pre-commit (Python/OCaml), lefthook (Go), cargo-husky (Rust). Use when the user wants commit-time formatting, linting, type-checking, or test gates. Detects ecosystem first.