Skillv1.0.1

ClawScan security

PopAI Powerpoint Slides · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

ReviewMar 13, 2026, 9:01 AM

Verdict: Review
Confidence: medium
Model: gpt-5-mini
Summary: The skill's code and instructions mostly match its stated purpose (creating PPTs via PopAI) but contain developer-specific paths, minor inconsistencies, and an undeclared dependency that warrant caution before installing.
Guidance: This skill appears to do what it claims (generate PPTs via PopAI) but has a few red flags you should address before use: 1) The SKILL.md run example uses a hard-coded developer path (/Users/Gunnar/...) — do not copy that verbatim; run the script from the skill directory or update the path. 2) The Python script requires the 'requests' library but the skill doesn't declare it; ensure your environment has requests installed (pip install requests) or the script will fail. 3) Uploaded local files are sent to PopAI's presigned S3 endpoint — only upload files you consent to share with PopAI. 4) Verify the POPAI_API_KEY you provide is scoped appropriately and not reused across unrelated services. 5) The slight S3 domain mismatch in documentation vs code (popai-file vs popai-file-boe) could be a harmless doc drift but verify returned download URLs point to PopAI's domains before trusting them. If you are not comfortable, run this in an isolated environment and inspect/run the included generate_ppt.py locally to confirm behavior. If anything looks unexpected, do not provide sensitive files or credentials.

Review Dimensions

Purpose & Capability: noteName/description, required env (POPAI_API_KEY), and network targets (api.popai.pro, popai-file S3) align with a PopAI PPT generator. However there are minor inconsistencies: SKILL.md final-result examples reference popai-file-boe.s3-accelerate.amazonaws.com while the script uploads to popai-file.s3-accelerate.amazonaws.com, and SKILL.md includes a hard-coded developer path (/Users/Gunnar/...) used in the example run command which is unrelated to the skill's declared scope.
Instruction Scope: concernInstructions ask the agent to run the included Python script in background and to poll a temp output file — behavior consistent with streaming generation. But the run example uses an absolute, developer-specific cwd (/Users/Gunnar/...), which is inappropriate and may cause failures or accidental path disclosure if copied literally. The instructions also tell the agent to read the output file via cat/Read (expected) but give the agent discretion to poll every ~30s; this is broad but understandable for streaming. Overall the scope is mostly appropriate, with an avoidable hard-coded path and somewhat permissive polling instructions.
Install Mechanism: noteNo install spec (instruction-only) — low-risk from installation perspective. The included script imports the 'requests' package but the skill does not declare this dependency; that may cause runtime failures if not present. Nothing in the manifest downloads arbitrary external code.
Credentials: okOnly POPAI_API_KEY is required and is the primary credential — this is proportionate for a service that calls the PopAI API and uploads files to PopAI-managed S3. No unrelated secrets or config paths are requested.
Persistence & Privilege: okalways is false and the skill is user-invocable (normal). The skill does not request elevated or persistent platform privileges beyond autonomous invocation (the platform default).