Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description (oral-history interviewing, structuring stories, generating short scripts) match the actual artifacts: helper questions, saving story segments to disk, and drafting scripts. No unrelated credentials, binaries, or installs are requested.
Instruction Scope
SKILL.md instructs the user to invoke the agent for interviews; it does not instruct the agent to read unrelated system files or environment variables or to send data to external endpoints. Note: the Python code saves records locally and exposes a small set of tool APIs (suggest_interview_questions, record_story_segment, transform_to_script). There is a minor code bug: get_interview_hooks falls back to hooks['通用'] but the hooks dict has no '通用' key, which can raise an error for unknown themes—this is a functionality issue, not a security one.
Install Mechanism
No install spec is provided (instruction-only plus a small code file). No external downloads, package installs, or archive extraction occur. This minimizes installation risk.
Credentials
The skill declares no required environment variables, credentials, or config paths and the code does not access environment secrets. This is proportionate to the skill's stated purpose.
Persistence & Privilege
The skill creates a local directory (default ./father_stories) and writes unencrypted JSON files named memory_YYYYMMDD_HHMMSS.json containing timestamps, content, tags, and status. It does not set 'always: true' and does not modify other skills. Consider privacy and file-permissions: these files may contain sensitive personal information and will persist on disk until removed.
Assessment
This skill appears to do what it says: provide interview prompts, save story segments, and generate script drafts. Before installing or running it, consider: 1) Where will files be written? The code uses a relative folder './father_stories' — if the agent runs in a shared or backed-up environment those JSON files could be exposed. Change the data directory to a safe, access-controlled path or review permissions. 2) Data retention and consent: these recordings/stories may contain personal or sensitive information — ensure you have consent and a deletion policy. 3) Code review: the included Python is small and has no network calls, but you may want to inspect or run it in a sandbox first. 4) Fixes: the get_interview_hooks fallback references a missing '通用' key (functionality bug) — you may want to patch that. If you need encryption, automatic upload, or integrations with social media/self-media platforms, expect the skill to require additional deliberate changes and credentials; currently it does not perform any external transmission.Like a lobster shell, security has layers — review code before you run it.
latestvk97d0z8ttxeq455r6dengwb9mn81s575
License
MIT-0
Free to use, modify, and redistribute. No attribution required.