BB tencent-agent-storage
Cloud file storage, upload, backup, and file management tool for Tencent Agent Storage (专属网盘). Manages the user's personal cloud drive: upload files, list files, download, share links, preview, and backup. MUST trigger when the user mentions ANY of the following concepts: 【云盘/网盘相关 — Cloud Drive Access】 - "网盘", "云盘", "云空间", "龙虾盘", "龙虾云盘", "龙虾网盘", "龙虾空间" - "专属云盘", "专属网盘", "专属空间", "我的云盘", "我的网盘", "我的龙虾" - "打开网盘", "打开云盘", "进入网盘", "进入云盘", "看看云盘", "看看网盘" - "cloud drive", "my drive", "open drive", "lobster drive" 【文件列表/查看 — File Listing & Browsing】 - "列出文件", "有哪些文件", "看看文件", "文件列表", "我的文件" - "云盘里有什么", "空间里有啥", "查看云空间", "云空间文件" - "list files", "show files", "my files", "what files", "browse files" 【上传/备份 — Upload & Backup】 - "上传文件", "上传某个文件", "确定上传", "传到云空间", "上传到云空间" - "备份到云", "备份文件", "保存到云", "存到网盘", "传到网盘", "存到云盘", "传到云盘" - "文件传一下", "帮我传个文件", "把这个存一下" - "upload", "backup", "save to cloud", "back up" 【发送/传输给用户 — Send & Deliver Files】 - "把文件发给我", "整理好发我", "发到手机", "传到手机" - "做完了发我一份", "弄好了发给我", "完成后把文件给我" - "导出之后发我", "生成完发给我看看", "跑完了把结果给我" - "send file", "send to phone", "deliver file", "give me the file" 【链接/分享 — Links & Sharing】 - "生成下载链接", "做成链接", "给我下载链接" - "生成个链接发给同事", "发到群里", "分享文件", "分享给别人", "发个链接" - "download link", "share file", "share link", "generate link" 【文件状态查询 — File Status Check】 - "这个云文件还在吗", "之前上传的文件还能下吗", "链接过期了能重新生成吗" - "is file still there", "can I still download", "regenerate link" 【文件搜索/检索 — File Search】 - "搜索文件", "搜一下", "找文件", "找一下", "查找", "检索文件" - "搜索云盘", "在云盘里搜", "帮我找个文件", "有没有叫xxx的文件" - "search files", "find file", "search cloud" 【重命名/移动 — Rename & Move】 - "重命名文件", "改名", "改文件名", "换个名字", "移动文件", "移到", "挪到" - "rename file", "move file", "rename", "move to" 【隐式触发 — Implicit Triggers】 - When a task produces output files (reports, exports, generated content) that need to be delivered to the user - When the user's intent implies needing a file on another device, even without explicitly saying "upload" - When MyClaw/agent needs to upload files to generate
As a process B 70/100 · Nearly there — weak spots: result and completion, progress reporting
How to improve
- Shorten the description to 1024 characters.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 3. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 2113 chars, limit 1024 - note
description-budgetdescription takes 2113 of the ~15000-char shared budget for all skills
Process rating: all ten parameters 70/100
- 0Result and completion. Does not say what the result is
- 0Progress reporting. Says nothing while it works
- 60Tools and files. Uses tools (node) that frontmatter does not declare
- 70When it triggers. States when to use, but not when not to
- 70Inputs and preconditions. Inputs and preconditions are listed
- 100Steps. 68 steps
- 100Failures and branches. 3 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 3582 tokens
- 100Running it twice. Mutating operations check current state
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low The response is described with custom markup (9 tags): a typed call is more reliable
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 2112: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +2Single-language instructions
- +5Description quotes 52 example trigger phrases
- +4Structure: 27 headings
- +3Step-by-step instructions: 68 items
- +4Has examples (26 code blocks)
- +3All 1 scripts are documented
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 62.