Escalation Brief Writer
v1.0.0把复杂问题整理成升级说明，减少来回追问，突出背景、影响、已尝试和诉求。；use for escalation, support, brief workflows；do not use for 省略关键信息, 把个人情绪写成事实.
⭐ 0· 109·0 current·0 all-time
byvx：17605205782@52yuanchangxing
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (escalation brief writer) align with files and code: resources/template.json, spec.json and scripts/run.py produce structured markdown briefs and lightweight audits. Required binary (python3) is appropriate and proportional.
Instruction Scope
SKILL.md confines behavior to reading local inputs, producing a reviewable draft or dry-run, and explicitly forbids executing high-risk commands or performing external writes without explicit boundaries. It suggests invoking scripts in-place or falling back to template-based output when execution isn't allowed.
Install Mechanism
No install spec; script is instruction-only plus a local Python script. No downloads, no external package installs, and the repo claims only Python standard library. This is low-risk.
Credentials
The skill requests no environment variables, credentials, or config paths. The script scans and reads files you supply; that file I/O is expected for a brief/ audit tool.
Persistence & Privilege
always is false, the skill is user-invocable and does not request persistent elevated privileges or modify other skills. Running the included script may write the specified output file but does not alter system or other skill configs.
Assessment
This skill appears coherent and low-risk: it reads local input files (or uses the template) and outputs structured escalation briefs. Before running, review any sensitive inputs you will pass (remove or redact PII/credentials), and inspect scripts/run.py if you want assurance it will only read/process the files you expect. The script does not perform network calls or require secrets, but it will read any path you give it — avoid pointing it at directories containing unrelated confidential data. If you enable actual execution inside an automated agent flow, ensure the agent is allowed only to call the script with vetted inputs.Like a lobster shell, security has layers — review code before you run it.
latestvk972k4zebtqh0dk8xwkmrwckes833an6
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🧯 Clawdis
OSmacOS · Linux · Windows
Binspython3