pp-shodhganga

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill instructs agents to replay playbook steps by substituting slots_resolved entries (which can include a "token" / "") into commands and also references stored secrets in credentials.toml, meaning the LLM may be required to include secret/token values verbatim in generated commands or output.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). The required workflow depends on Shodhganga/HTTP HTML retrieval (“only HTML”) during commands like harvest/thesis/browse, so outsider-authored web page text can be ingested by the CLI and then passed back to the agent/LLM via the JSON/agent output channel.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs installing the CLI by fetching and executing remote code (e.g., go install github.com/mvanhorn/printing-press-library/library/productivity/shodhganga/cmd/shodhganga-pp-cli@latest and go install github.com/mvanhorn/printing-press-library/library/productivity/shodhganga/cmd/shodhganga-pp-mcp@latest), which are runtime external fetches that execute remote code and the skill requires the CLI to operate.