ceo-mail-review
Skill0 starsChanged 2 months agoMITView on GitHub
Use for reviewing an incoming mail card, resolving the complete message or thread, inspecting attachments and links, checking whether a reply already exists, and drafting or sending an authorized reply. Use ceo-document-review for a standalone material review outside the mail workflow. Load dingtalk-mail before any DingTalk mail operation.
In this repository42 other files it gives its agents.
- ai-pr-redteamSkill
- ai-servicesSkill
- build-work-memorySkill
- ceo-calendar-inviteSkill
- ceo-document-reviewSkill
- ceo-email-classifierSkill
- ceo-feedback-iterationSkill
- ceo-feedback-processingSkill
- ceo-meeting-workSkill
- ceo-message-triageSkill
- ceo-minutes-syncSkill
- ceo-personnel-communicationSkill
- ceo-wechatSkill
- ceo-work-trackingSkill
- codex-model-capacity-retrySkill
- daily-frontier-tech-discoverySkill
- dengbao-code-auditSkill
- dingtalk-browser-exportSkill
- dingtalk-knowledge-organizeSkill
- dingtalk-minutes-access-requestSkill
- dingtalk-oa-approvalSkill
- dingtang-okr-reviewSkill
- fundflow-investor-meetingSkill
- fxiaoke-crm-cliSkill
- internal-app-standardsSkill
- ocrSkill
- pm-data-adaptation-assistantSkill
- project-delivery-document-collectionSkill
- qa-generated-test-caseSkill
- senior-technical-product-evaluationSkill
- spec-intakeSkill
- sso-oidc-onboardingSkill
- stardust-interviewSkill
- production-devops-sreSkill
- stardust-ttsSkill
- startask-io-contract-auditSkill
- task-managementSkill
- transcribeSkill
- veyra-timesheetSkill
- vibe-coding-rescueSkill
- vibe-codingSkill
- video-transcribeSkill
The file
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 7 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 16
Use the platform's mail and material Skills instead of copying their commands into this business workflow.
- Distinctive: line 32
Treat a truncated card or quoted preview only as a locator.
- Distinctive: line 40
Check the current thread, sent state, and safe prior receipts before proposing a reply.
- Distinctive: line 46
Do not infer or invent unread content.
- Very specific: line 57
If the complete current thread shows an equivalent reply already sent, use canonical no_action for the mail effect and report the verified …
- Distinctive: line 60
If review is complete but reply authorization is absent, provide only the requested review or draft through an authorized channel.
- Distinctive: line 63
If evidence access fails, report the dependency failure rather than drafting from the preview.
- Distinctive: line 66
When a participant can resolve a genuine evidence gap, ask one concrete question naming the specifically missing mail or linked material.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
CEO Mail Review
Load ceo-mail-review for mail review, thread resolution, linked-material
inspection, reply judgment, and authorized reply proposals.
Compose Operation Skills
Use the platform's mail and material Skills instead of copying their commands into this business workflow.
| Source | Operation Skills |
| DingTalk mail | dingtalk-mail |
| Lark mail | lark-mail |
| DingTalk linked material | dingtalk-doc, dingtalk-aitable, or dingtalk-drive |
| Lark linked material | lark-doc, lark-base, or lark-drive |
Load ceo-document-review when the requested outcome includes a substantive
review of linked material. Load the matching chat Skill only when the result
must also be delivered in the source conversation.
Resolve Complete Evidence
- Treat a truncated card or quoted preview only as a locator. Resolve the principal's mailbox and the complete original message or thread with the loaded mail Skill.
- Confirm sender, recipients, subject, current thread state, and the exact request. Do not ask the sender to paste content that the loaded mail Skill can read.
- Inspect every linked material needed for the requested judgment with its matching operation Skill. A link title or mail summary is not its content.
- Check the current thread, sent state, and safe prior receipts before proposing a reply. Do not propose or execute a duplicate reply.
The agent performs the business judgment. The service supplies references and exact commands without interpreting mail or linked content. Use supplied exact commands as given and load operation Skills for any additional reads or writes. Do not infer or invent unread content.
Authorization And Outcome
Propose a mail reply only when the current request explicitly authorizes replying. Review-only, summarize-only, or approval-only requests do not authorize a mail reply. Authorization from an older message does not silently carry into a materially different current request.
- A justified mail reply is one action in a canonical
proposal; the mail operation Skill owns execution and exact readback. - If the complete current thread shows an equivalent reply already sent, use
canonical
no_actionfor the mail effect and report the verified state only when the current conversation needs it. - If review is complete but reply authorization is absent, provide only the requested review or draft through an authorized channel; do not execute or propose a mail send.
- If evidence access fails, report the dependency failure rather than drafting from the preview.
When a participant can resolve a genuine evidence gap, ask one concrete question naming the specifically missing mail or linked material. Do not ask for a generic resend or for content the loaded operation Skills can retrieve.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

