grant-proposal
Draft a structured grant proposal from research ideas and literature. Supports KAKENHI (Japan), NSF (US), NSFC (China, including 面上/青年/优青/杰青/海外优青/重点), ERC (EU), DFG (Germany), SNSF (Switzerland), ARC (Australia), NWO (Netherlands), and generic formats. Use when user says "write grant", "grant proposal", "申請書", "write KAKENHI", "科研費", "基金申请", "写基金", "NSF proposal", or wants to turn research ideas into a funding application.
70
—
Does it follow best practices?
Impact
—
No eval scenarios have been run
Advisory
Suggest reviewing before use
Security
3 findings — 3 medium severity. This skill can be installed but you should review these findings before use.
W011: Third-party content exposure detected (indirect prompt injection risk)
What this means
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
Why it was flagged
Third-party content exposure detected (high risk: 0.85). In Phase 1 the skill invokes `/research-lit "$ARGUMENTS"`, which performs multi-source literature search and WebSearch for funded projects (public web content fetched at runtime), and that fetched free-form page text can be ingested into the agent/LLM context for drafting and gap/landscape summaries.
W012: Unverifiable external dependency detected (runtime URL that controls agent)
What this means
The skill fetches instructions or code from an external URL at runtime, and the fetched content directly controls the agent’s prompts or executes code. This dynamic dependency allows the external source to modify the agent’s behavior without any changes to the skill itself.
Why it was flagged
Potentially malicious external URL detected (high risk: 0.90). The skill accepts a --style-ref source that can be an http(s) URL (e.g., an arXiv or other web URL) which is fetched at runtime via the STYLE_HELPER (python3 "$STYLE_HELPER" --source "<source>") and the resulting style_profile.md is used to steer the agent's proposal-structure prompts, so remote http(s) sources passed to --style-ref can directly control agent behavior.
W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection)
What this means
Detected hidden or invisible Unicode characters (Format/Cf or Control/Cc categories) in the component’s content. These characters are invisible when rendered but are still processed by AI models, and attackers use them to smuggle instructions past human review — for example, zero-width spaces, bidirectional overrides, invisible formatters, or Unicode Tag characters (U+E0000–U+E007F) that encode an entire hidden message. Severity escalates to high when three or more distinct hidden character types are present, or when a hidden tag-encoded message is successfully decoded, as these strongly indicate intentional obfuscation.
Why it was flagged
Hidden Unicode characters detected (1 type(s) found)
- Commit
fe5963c
- Audited
- Security analysis
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.