Nonprofit Grant Outline
v1.0.0为公益项目生成 grant proposal 大纲，连接目标、预算、影响指标与证据。；use for nonprofit, grant, proposal workflows；do not use for 伪造受益人数, 替代正式合规审核.
⭐ 0· 105·0 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 (grant proposal outlines) match the included resources: template, spec.json, examples, README, and a local Python script (scripts/run.py) that produces structured Markdown reports. Required binary (python3) is appropriate and nothing else is requested.
Instruction Scope
SKILL.md instructs the agent to prefer using the bundled script when shell/exec is available and otherwise to use the local template/spec. The script itself supports modes beyond simple summarization (directory scanning, CSV auditing, pattern scanning, and a skill-audit mode) and will read files you point it at. The default spec.json mode is 'structured_brief' (which aligns with the stated purpose), but the script can enumerate and read many file types if invoked with directory or file inputs—so avoid passing directories containing secrets or unrelated system data.
Install Mechanism
No install spec or remote downloads; this is instruction-only with a local Python script relying only on the standard library. No network fetches or extract/install steps are present.
Credentials
The skill requires only python3 and declares no environment variables, credentials, or config paths. There are no disproportionate secret requests.
Persistence & Privilege
always is false, the skill is user-invocable and not forced on all agents. It does not modify other skills or request persistent system-wide changes. The script may write an output file if you pass --output, which is expected behaviour.
Assessment
This skill appears to do what it says: assemble grant-outline drafts from inputs and local templates. Before running: (1) inspect scripts/run.py (it reads files/directories you point it at and can perform pattern/dir audits), (2) do not pass folders containing secrets or system config as the --input, (3) prefer using --dry-run first and run in an isolated/sandboxed workspace, and (4) verify outputs manually (the skill explicitly warns against fabricating beneficiary counts or replacing formal compliance review). If you need networked lookups or integration with external systems, treat those as separate steps and do not supply credentials to this skill.Like a lobster shell, security has layers — review code before you run it.
latestvk9700yqzwgv7yk2trfr117jq49837zc0
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🌱 Clawdis
OSmacOS · Linux · Windows
Binspython3