comet-build

仅在用户明确调用 /comet-build，或由 Comet 根 Skill/runtime 路由到 full workflow 的 build 阶段时使用；创建或恢复实施计划并执行任务。

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

77%Weight 40%Scale 1-3

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

A highly actionable, well-sequenced orchestration skill with strong validation checkpoints, but it is verbose with repeated prohibitions and presents as a monolithic document that does not use progressive disclosure to split out its detailed reference material.

Suggestions

Deduplicate the repeated guard/prohibition text for each state field (isolation, subagent_dispatch, tdd_mode, review_mode) into a single shared statement, and consolidate the multiple re-statements of the Step 2 联合决策 point.

Move the detailed review-gate rules and the isolation/execution/TDD/review option tables into reference files under references/, keeping SKILL.md a lean overview with one-level-deep, clearly signaled links.

Pull the dated branch-naming example ('今天是 2026-06-09') out of inline prose into a parameterized note so time-sensitive specifics do not clutter the core workflow.

Dimension	Reasoning	Score
Conciseness	The body is mostly actionable commands and tables rather than concept explanation, but it is heavily padded: the same "联合决策" point and the guard-failure prohibition for each state field are restated multiple times, and inline dated examples could be tightened. It respects Claude's competence but repeats directives extensively.	2 / 3
Actionability	It provides concrete, copy-paste-ready commands throughout (`comet state set/select/check`, `git checkout -b`, `git rev-parse HEAD`, `comet guard ... build --apply`) plus specific option tables for isolation, execution, TDD and review modes.	3 / 3
Workflow Clarity	The build phase is sequenced as Steps 0–5 with explicit validation checkpoints (Entry Check, BLOCKED handling, build→verify guard) and feedback loops (systematic-debugging on failure, context-recovery on resume), plus a clear exit-conditions checklist for risky state-transition operations.	3 / 3
Progressive Disclosure	No bundle files exist, so the SKILL.md is a monolithic wall of text with large inline decision tables and review-gate detail that could be split out; its `comet/reference/*.md` links are external runtime references rather than skill-bundle files, so it is not truly offloading detail one level deep.	2 / 3
	Total	10 / 12 Passed

Description

90%Weight 40%Scale 1-3

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 tight, well-scoped description with an explicit user trigger and clear what/when coverage. Its only weakness is that the action list is high-level rather than enumerating the specific build-phase operations.

Dimension	Reasoning	Score
Specificity	It names the domain (build phase) and concrete actions — "创建或恢复实施计划并执行任务" (create or resume implementation plan and execute tasks) — but stops at a high-level summary without enumerating the multiple specific actions the body actually performs (state management, isolation, TDD, review).	2 / 3
Completeness	It explicitly answers both questions: "when" via "仅在用户明确调用 /comet-build，或由 Comet 根 Skill/runtime 路由到 full workflow 的 build 阶段时使用" and "what" via "创建或恢复实施计划并执行任务".	3 / 3
Trigger Term Quality	The explicit slash-command trigger "/comet-build" is exactly what a user would naturally say to invoke the skill, and it is reinforced by the "full workflow 的 build 阶段" routing phrasing.	3 / 3
Distinctiveness Conflict Risk	Scoped tightly to the /comet-build command and the Comet build phase, giving it a clear niche distinct from sibling phase skills (open/design/verify) with little overlap risk.	3 / 3
	Total	11 / 12 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: assets/skills-zh/comet-build/SKILL.md
Commit: ea9eefe

Reviewed: 8 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.