CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

spek-new

Create a new Specification for a feature, from a conversation or from existing material such as an issue, a tracker epic, a design document, a file, a web page or pasted text ("spec from

Quality

90%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

85%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

An exceptionally actionable, well-gated workflow document — every command is copy-paste ready and every branch has an explicit error path and user decision point. Its weaknesses are repetition of the anti-stop warning and a total absence of progressive disclosure: ~290 lines inlined in a single file with material (the split flow above all) that belongs in one-level-deep reference files.

Suggestions

Move the split check and split flow (~100 lines) into a references/split.md loaded only when a split is requested, keeping SKILL.md an overview with a clearly signaled pointer.

Consider the same treatment for the design-document section and resume details, which are only needed on their respective paths.

Consolidate the three restatements of the 'do not stop after spec new' warning in the opening into one authoritative statement to trim token cost without losing the emphasis.

DimensionReasoningScore

Conciseness

Nearly every sentence is CLI-specific operational instruction with no generic concept explanations, but there is noticeable repetition: the "do not stop after spec new" warning appears three times in the opening (the STOP block, "This is a loop", and "Concretely: do not stop after spec new"), and split-decline handling is stated twice ("On a decline, continue exactly as if no split had been offered" / "On a decline, nothing is written"). Between anchors 3 and 4, noticeably above the midpoint.

4 / 5

Actionability

Fully executable throughout: exact commands with complete --data JSON payloads (spec new with sources/epic, epic write/epic split staged under .spektacular/tmp/ with scratch-file cleanup), per-error-code responses (uncommitted_changes, auto_commit_failed, epic_complete), and a concrete epic_split.json outline. Copy-paste ready.

5 / 5

Workflow Clarity

The state-machine loop is explicit (read instruction → perform → goto → repeat until the CLI says finished) with validation gates at every decision point: version check first, resume check before naming, the uncommitted-changes gate routed to the user, epic-complete confirmation, subagent review of split results, and the atomic restore on partial split failure. Feedback loops for error recovery are present throughout.

5 / 5

Progressive Disclosure

No bundle files exist, so the entire 290-line body is one monolithic SKILL.md: the ~100-line split check/flow, design-document handling, and resume logic are all inlined where they clearly belong in separate reference files. Section headers are clear and well-organized (above anchor 2's 'minimal structure'), matching anchor 3's 'content that should be separate is inline'.

3 / 5

Total

17

/

20

Passed

Description

91%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description: concrete, comprehensive over entry modes, and enriched with verbatim example trigger phrasings that make natural-language matching easy. The only gap is the absence of an explicit 'Use when…' clause framing the triggers.

Suggestions

Append an explicit trigger clause, e.g. "Use when the user asks to create, draft, or start a spec/specification, or to split an already-written spec into an epic", to lift completeness to the anchor-5 formulation.

DimensionReasoningScore

Specificity

Enumerates multiple concrete actions and every entry mode: "Create a new Specification for a feature, from a conversation or from existing material such as an issue, a tracker epic, a design document, a file, a web page or pasted text … or split an already-written spec into an epic." Coverage is comprehensive with no gaps among the skill's entry points.

5 / 5

Completeness

The 'what' is clear (create a spec, or split a written spec into an epic) and 'when' is conveyed through explicit example trigger phrasings, but there is no 'Use when…' clause or equivalent framing. Anchor 4 fits: both present, with the 'when' slightly less explicit than the anchor-5 formulation.

4 / 5

Trigger Term Quality

Embeds natural user phrasings verbatim — "spec from #45", "start a spec from LIN-123", "a pasted link" — plus synonyms (spec/Specification, epic, split) covering how users actually request this. No common variation is obviously missing.

5 / 5

Distinctiveness Conflict Risk

Clear niche — specification creation in the Spektacular toolchain — with distinct triggers ('spec', 'Specification', 'epic', 'split') that are unlikely to fire for a different skill.

5 / 5

Total

19

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

nicholasjackson/tcgreplay

Path

skills/spek-new/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.