CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

fastgpt-backport-pr

将 FastGPT 已合入的 GitHub PR cherry-pick(pick、回补、backport)到用户明确提供的维护分支,验证后直接推送并对齐维护版 Pro 子模块;随后更新对应进行中版本的中英文文档,单独向 upstream/main 提交文档 PR。不用于合并源 PR 或执行发布。

Quality

81%

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

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.

A rigorous, highly actionable workflow with strong sequencing, validation checkpoints, and error-recovery loops for risky git operations. Its weaknesses are verbosity from repeated caveats across sections and a monolithic single-file structure where the submodule and documentation procedures would benefit from being split into reference files.

Suggestions

Deduplicate repeated caveats: state the SHA-preservation limitation and the no-empty-commit rule once each and reference them, instead of repeating them in sections 4, 5, and 6.

Move the Pro submodule alignment procedure (section 4) and the version-docs workflow (section 6) into separate reference files (e.g., references/pro-submodule.md, references/docs-pr.md) with clearly signaled links from SKILL.md.

Spell out the doc validation commands (Prettier, textlint, checkEnglishDocs invocation) or point to the exact script(s) so the documentation phase is as copy-paste-ready as the git phases.

DimensionReasoningScore

Conciseness

The body is entirely operational policy with no explanation of concepts Claude already knows, but it is noticeably padded: the SHA-preservation caveat appears in the intro bold block and again in sections 4 and 5, "不创建空提交" is repeated three times, and many long conditional sentences could be tightened. This fits "mostly efficient but could be tightened" better than the repeated-unnecessary-explanation of the 2 anchor.

3 / 5

Actionability

Core operations are given as concrete, executable commands with proper quoting — gh api with pagination, git cherry-pick -x -m <parent-number>, git push --force-with-lease with an explicit expected remote SHA. Minor gaps remain: the doc-section validation (Prettier, textlint, checkEnglishDocs) names checks but defers exact commands, keeping it just below fully copy-paste-ready.

4 / 5

Workflow Clarity

Six clearly sequenced sections with explicit validation checkpoints — verifying base_sha..HEAD, git diff --check, submodule gitlink comparison — and genuine feedback loops for the risky batch/destructive operations: logical conflicts pause for user decision, lease rejection triggers one bounded resync, and the remote SHA is re-verified after push. This matches the clear-sequence-with-explicit-validation anchor.

5 / 5

Progressive Disclosure

The body is well-sectioned and external links are clearly signaled, but no bundle files exist and roughly 150 lines of dense policy are all inlined — the Pro submodule procedure (section 4) and the docs-update workflow (section 6) are prime candidates for separate reference files. This matches "some structure but content that should be separate is inline"; it is above the 2 anchor because structure is thorough, not minimal.

3 / 5

Total

15

/

20

Passed

Description

82%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 highly specific, distinctive description with excellent natural-synonym trigger terms (cherry-pick/pick/回补/backport) and comprehensive concrete actions. Its main weakness is the absence of an explicit "Use when…" invocation clause, leaving the "when" only weakly implied and capping completeness at 3.

Suggestions

Add an explicit trigger clause such as "Use when the user asks to cherry-pick/backport already-merged FastGPT PRs to a maintenance branch" to answer 'when' directly and lift completeness.

State the invocation condition positively (e.g., user supplies PR number/URL plus target branch) rather than only listing what the skill is not for.

Consider trimming the single run-on sentence into two shorter sentences so the actions and the boundary statement each read cleanly.

DimensionReasoningScore

Specificity

The description enumerates multiple concrete actions covering the entire workflow — "cherry-pick(pick、回补、backport)到…维护分支", "验证后直接推送", "对齐维护版 Pro 子模块", "更新…中英文文档", "提交文档 PR" — with no vague filler, matching the comprehensive-coverage anchor. Score 4 would require minor gaps in coverage, but every phase of the task is explicitly named.

5 / 5

Completeness

The "what" is clear and detailed, but there is no "Use when…" clause or equivalent explicit trigger guidance — the "when" is only weakly implied by the verb synonyms, and the negative boundary ("不用于合并源 PR 或执行发布") states exclusions rather than invocation conditions. Per the rubric guideline, a missing explicit trigger clause caps completeness at 3.

3 / 5

Trigger Term Quality

It explicitly enumerates the natural synonyms a user would say — "cherry-pick", "pick", "回补", "backport" — alongside "GitHub PR", "维护分支", and "文档", giving comprehensive natural-term coverage. The 4 anchor ("a few natural terms missing") fits worse since the common variations are all present.

5 / 5

Distinctiveness Conflict Risk

The description occupies a clear niche — backporting merged FastGPT PRs to maintenance branches plus a separate docs PR to upstream/main — with distinct triggers, so conflict with other skills is minimal. It is far more specific than the 4 anchor's "minor overlap risk with closely related skills".

5 / 5

Total

18

/

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

labring/FastGPT

Path

.agents/skills/system/fastgpt-backport-pr/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.