comet-design

Comet Phase 2: Deep Design. Invoke with /comet-design. Produce Design Doc and delta spec through brainstorming.

Quality

61%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./eval/local/skills/benchmarks/039-release/comet-classic-039-design/SKILL.md

Quality

Content

70%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.

The content is a well-sequenced, highly actionable complex workflow with strong validation checkpoints, feedback loops, and an exit checklist. Its main weaknesses are repeated guidance that bloats the token budget and inlined detail plus external (non-bundled) references that limit progressive disclosure.

Suggestions

De-duplicate repeated guidance — state the Design Doc frontmatter and compaction-recovery procedure once and cross-reference them, instead of restating in Steps 1b, 1d, 1e, and 2.

Move the handoff-package format spec and the brainstorm-summary template into bundled reference files (e.g. references/handoff-format.md, references/brainstorm-summary.md) and link to them from the body.

Clarify that `comet/reference/decision-point.md`, `context-recovery.md`, and `auto-transition.md` are external comet-skill files, or bundle the relevant excerpts so the skill is self-contained.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient operational detail for a genuinely complex workflow, but repeats the same guidance in multiple places (Design Doc frontmatter shown in both Step 1b and Step 2, compaction-recovery notes restated across 1d/1e/Step 2, 'do not write the Design Doc early' restated several times); this fits the 'mostly efficient but could be tightened' anchor rather than the minor-trim level 4.	3 / 5
Actionability	Every step ships concrete, copy-paste-ready bash commands (entry verification, handoff generation, state/guard invocations) and a brainstorm-summary.md template, but relies on placeholder substitution (<name>, <change-name>) and delegates the core design work to the external Superpowers brainstorming skill, leaving minor gaps short of fully-executable level 5.	4 / 5
Workflow Clarity	The workflow is explicitly sequenced (Steps 0, 1a–1e, 2, 3) with concrete validation steps (Step 0 entry verification, the blocking user-confirmation gate in 1c, `guard --apply` at exit), feedback loops (compaction-recovery resume, 'return to Step 1b/1c if not confirmed'), and an Exit Conditions checklist — matching the explicit-validation + feedback-loops + checklist anchor.	5 / 5
Progressive Disclosure	No bundle files exist (references/scripts/assets absent), yet the ~260-line body inlines content that could live in separate reference files (the handoff-package spec, the brainstorm-summary template, context-recovery details) and references external `comet/reference/*` paths that are not part of this bundle and not clearly signaled as navigable references; this fits the some-structure / content-could-be-separate anchor better than the well-split level 4.	3 / 5
	Total	15 / 20 Passed

Description

53%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.

The description is reasonably specific and distinct, naming concrete deliverables and a clear slash-command trigger, but it lacks an explicit situational 'Use when...' clause and broad natural-keyword coverage. It is solid mid-tier: clear on 'what', weak on 'when'.

Suggestions

Add an explicit situational trigger, e.g. 'Use when starting the design phase of a Comet change or when the user asks to produce a Comet Design Doc'.

Broaden natural keywords/synonyms (e.g. 'technical design', 'architecture', 'spec patches') so users phrasing the need differently still match.

Name a third concrete action (e.g. 'write back Spec Patches to OpenSpec delta specs') to lift specificity toward comprehensive coverage.

Dimension	Reasoning	Score
Specificity	"Produce Design Doc and delta spec through brainstorming" names the domain (Comet Phase 2 design) and two concrete deliverables, matching the 'names domain and 1-2 concrete actions' anchor; it is not comprehensive enough for a 4 and has more concrete actions than the generic 'Processes PDF files' level-2 anchor.	3 / 5
Completeness	The 'what' is clear (produce a Design Doc and delta spec via brainstorming), but the 'when' is only weakly implied via the slash-command invocation rather than a situational 'Use when...' clause; this fits the clear-what / weak-when anchor and does not reach the explicit-when level 4.	3 / 5
Trigger Term Quality	"Invoke with /comet-design" plus "brainstorming" and "Design Doc" provide some relevant natural keywords, but common variations/synonyms are missing; coverage is above the one-or-two-generic-keywords level 2 but below the good-coverage level 4.	3 / 5
Distinctiveness Conflict Risk	"Comet Phase 2: Deep Design" with the distinct "/comet-design" trigger carves a clear niche, but it has minor overlap risk with closely related comet phases and the brainstorming skill it delegates to, placing it just below the minimal-conflict level 5.	4 / 5
	Total	13 / 20 Passed

Validation

100%

Warnings & errors only

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: rpamis/comet
Path: eval/local/skills/benchmarks/039-release/comet-classic-039-design/SKILL.md
Commit: 899d1fb

Reviewed: 7 days ago

Table of Contents

Discovery Implementation Validation

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.