clickupfy-release
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from Git commit history and PR descriptions to automate release versioning and changelog generation.
- Ingestion points: Git commit messages (Conventional Commits),
.release-please-manifest.json, and GitHub Release PR content (SKILL.md). - Boundary markers: The instructions rely on specific prefixes (fix, feat) but do not define explicit delimiters or instructions to ignore malicious content embedded within commit messages.
- Capability inventory: Executes npm scripts (
npm ci,npm run validar,npm run build:executable) and runs generated local binaries (SKILL.md). - Sanitization: No specific sanitization or filtering logic is defined for the agent to handle potentially malicious commit messages.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands for building, validating, and testing releases locally.
- Evidence: Commands such as
npm run validar,npm run release:check, and direct execution of built artifacts like./artifacts/clickupfy-v0.1.0-linux-x64/clickupfy(SKILL.md). - [EXTERNAL_DOWNLOADS]: The skill involves downloading dependencies and interacting with external registries.
- Evidence: Usage of
npm cito install project dependencies and publishing packages to the npm registry (SKILL.md). These operations target well-known and expected services.
Audit Metadata