proactive-send

向 awada 客户主动发送消息。在 openclaw 消息处理循环之外直接 POST 到 relay 网关 outbound 端点，无需等待客户发起对话。

Quality

68%

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

Fix and improve this skill with Tessl

tessl review fix ./crews/sales-cs/skills/proactive-send/SKILL.md

Quality

Content

87%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 lean, highly actionable, and well-structured for a simple single-purpose skill, with a verified bundle script and a clearly signaled one-level reference. The only gap is the absence of an explicit pre-send verification checkpoint for an outward-facing, hard-to-reverse operation, which caps workflow clarity.

Suggestions

Add an explicit pre-send validation checkpoint (e.g., confirm recipient and message content before invoking, or a dry-run/--dry-run flag) since proactive customer messages are outward-facing and hard to reverse.

Make the link to the transport contract an explicit clickable reference (e.g., 'See [docs/AWADA-CLIENT-TRANSPORT.md](docs/AWADA-CLIENT-TRANSPORT.md) §3') so the one-level reference is clearly signaled as navigation rather than inline prose.

Optionally surface the underlying script invocation (node scripts/send.mjs ...) alongside the proactive-send wrapper so the executable path is unambiguous when the wrapper is unavailable.

Dimension	Reasoning	Score
Conciseness	Lean body with no concept-explanation padding; the parameter table and terse '注意事项' bullets assume Claude's competence and every line earns its place. Not below 5 because nothing is over-explained.	5 / 5
Actionability	Provides a copy-paste-ready command, a parameter table with concrete value sources ('对话上下文 Sender 块的 id 字段'), auto-loaded config note, and an explicit return contract (prints streamId like '1234-0', exit 0/1). Covers the common case fully.	5 / 5
Workflow Clarity	Single-action skill with an error-feedback loop (stderr + exit 1 on HTTP failure), but sending a real customer message is outward-facing/irreversible and lacks an explicit pre-send verification checkpoint; the destructive-operations cap overrides the simple-skill exception. Not 4 because validation checkpoints are implicit, not 2 because a rough sequence and error feedback are present.	3 / 5
Progressive Disclosure	Under-50-line skill with clean sections (使用方法/参数说明/返回值/注意事项), a real bundle script (scripts/send.mjs), and a single well-signaled one-level reference (docs/AWADA-CLIENT-TRANSPORT.md §3). Meets the simple-skill 5-anchor for well-organized structure.	5 / 5
	Total	18 / 20 Passed

Description

50%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 conveys a specific, distinguishable capability but leans on internal jargon and omits an explicit 'Use when' trigger clause, weakening trigger-term quality and completeness. It is clear on the 'what' but weak on the 'when'.

Suggestions

Add an explicit 'Use when ...' trigger clause naming natural situations (e.g., 'Use when sales-cs needs to proactively follow up with an awada customer outside an active conversation').

Soften internal jargon ('openclaw 消息处理循环', 'relay 网关 outbound 端点') or pair it with user-facing phrasing so natural trigger terms are present alongside the technical detail.

Include natural synonyms users might say (e.g., '主动发消息', '主动联系客户', 'push 消息') to broaden trigger-term coverage.

Dimension	Reasoning	Score
Specificity	Names the domain (awada outbound) and 1-2 concrete actions ('主动发送消息', '直接 POST 到 relay 网关 outbound 端点'), but coverage is not comprehensive. Not 4 because only ~2 actions are listed, not several; not 2 because it goes beyond merely naming the domain with a concrete POST action.	3 / 5
Completeness	Clearly states the 'what' (proactive send to awada customer) but lacks an explicit 'Use when...' trigger clause; 'when' is only weakly implied via '无需等待客户发起对话'. Capped at 3 per the missing-trigger-clause guideline.	3 / 5
Trigger Term Quality	Contains one natural phrase ('主动发送消息', 'awada 客户') but is dominated by internal-system jargon ('openclaw 消息处理循环', 'relay 网关 outbound 端点') that users would not naturally utter. Below the midpoint because jargon outweighs natural trigger terms.	2 / 5
Distinctiveness Conflict Risk	Occupies a clear, narrow niche (proactive outbound via relay gateway outside the openclaw loop) with distinct triggers and minimal overlap risk with other skills.	5 / 5
	Total	13 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning

	Total	14 / 16 Passed

Repository: TeamWiseFlow/xiaobei
Path: crews/sales-cs/skills/proactive-send/SKILL.md
Commit: fdc03d6

Reviewed: 11 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.