social-push

使用 agent-browser 帮用户将内容发到社交媒体上。当用户需要发布内容、推送文章、上传文章、发帖到社交平台时使用此 skill。

Quality

80%

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 ./app/skills/social-push/SKILL.md

Quality

Content

82%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 body is a lean, actionable overview with concrete agent-browser commands, explicit per-step validation, a fix-and-verify recovery loop, and well-organized per-platform references. Remaining gaps are the ambiguous browser-launch command, a grammatically tangled add-new-platform procedure, and unreferenced scripts/assets in the bundle.

Suggestions

Fix the browser-launch rule to two unambiguous commands, e.g. `open -na "Google Chrome" --args --remote-debugging-port=9222` (and the Edge equivalent), instead of the "Chrome / Edge" slash placeholder.

Rewrite "添加新的社交平台" steps 3–4 as a single clear conditional: test the full interaction path step by step in a launched browser, and only after every step is verified create the new workflow file and add its link under References.

Reference the bundle's scripts/ (render_xhs* for image rendering) and assets/ (themes, example.md) from the relevant sections so all bundle content is discoverable.

Dimension	Reasoning	Score
Conciseness	The ~60-line body is lean and assumes Claude's competence: rules, workflow, and self-evolution steps carry only operational content with zero padding or explanations of concepts Claude already knows. Every section earns its tokens, matching anchor 5; there are no unnecessary explanations that would drop it to anchor 4.	5 / 5
Actionability	Concrete executable commands are given throughout (`open -na "Google Chrome" … --remote-debugging-port=9222`, `agent-browser --auto-connect`, `agent-browser snapshot -i`, `agent-browser eval "js"`) plus an explicit AskUserQuestion step. Minor gaps keep it at anchor 4: the browser-launch command is an ambiguous Chrome/Edge placeholder rather than a copy-paste-ready invocation.	4 / 5
Workflow Clarity	The Core Workflow is a clear 5-step sequence with an explicit per-step validation rule ("每步操作后用 agent-browser snapshot -i 确认元素 ref") and a fix-and-verify feedback loop in Self-evolution. It falls short of anchor 5 because the "添加新的社交平台" steps 3–4 form a tangled conditional sentence and the browser-launch step is ambiguous, leaving minor clarity gaps.	4 / 5
Progressive Disclosure	The body is an overview with well-signaled, one-level-deep references organized by platform with markdown links — matching the anchor-5 pattern for references. However, the bundle's scripts/ (render_xhs*.js/py) and assets/ (themes, card.html, cover.html) are never mentioned in SKILL.md, leaving orphaned content and a minor organization gap at anchor 4.	4 / 5
	Total	17 / 20 Passed

Description

70%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 answers both "what" and "when" with natural trigger phrases and a clear tool-mediated action, avoiding vagueness. Its main weaknesses are a generic capability statement and the absence of the supported platform names, which would improve both specificity and trigger coverage.

Suggestions

Enumerate concrete capabilities in the description, e.g. "上传文章和图片、填写标题与话题标签、暂存草稿" (upload articles and images, fill in titles and hashtags, save drafts), to raise specificity.

Include the supported platform names (小红书, 微博, X, 微信公众号, 掘金, Linux.do, 知乎) as trigger terms so users who name a platform land on this skill.

State the draft-only safety behavior (never auto-publishing) in the description to sharpen the "what" and further reduce overlap with content-writing skills.

Dimension	Reasoning	Score
Specificity	Names the domain and tool ("使用 agent-browser 帮用户将内容发到社交媒体上") with one clear action, but does not enumerate concrete capabilities such as image upload, draft saving, or the per-platform posting the skill actually supports. This matches anchor 3 (domain plus 1-2 concrete actions, not comprehensive); it is above anchor 2 because the action and tool are explicit, and below anchor 4 because no list of several specific actions is given.	3 / 5
Completeness	Both parts are present: a "what" (use agent-browser to post content to social media) and an explicit "when" clause with concrete trigger phrases ("当用户需要发布内容、推送文章、上传文章、发帖到社交平台时使用此 skill"). It sits at anchor 4 rather than 5 because the "what" is generic and the "when" omits platform-specific triggers.	4 / 5
Trigger Term Quality	The trigger phrases "发布内容、推送文章、上传文章、发帖到社交平台" are natural terms a user would say, giving good keyword coverage. It falls short of anchor 5 because platform names (小红书, 微博, X/Twitter, 微信公众号, 知乎, 掘金) — which users would naturally mention — are absent.	4 / 5
Distinctiveness Conflict Risk	Social-media publishing via agent-browser is a clear niche with distinct triggers (发帖, 推送文章, 社交平台), so conflict risk is low. Minor overlap remains with content-writing skills since the description does not name specific platforms, keeping it at anchor 4 rather than 5.	4 / 5
	Total	15 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning

	Total	15 / 16 Passed

Repository: ZHangZHengEric/Sage
Path: app/skills/social-push/SKILL.md
Commit: 81bddde

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.