plan-alm

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill acquires access tokens and then shows them being interpolated directly into CLI args and HTTP Authorization headers (e.g. --token "{DEV_TOKEN}", --bapToken "{BAP_TOKEN}", Authorization: Bearer {DEV_TOKEN}), which requires the LLM to handle or emit secret values verbatim.

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

• Third-party content exposure detected (medium risk: 0.30). In plan-alm’s runtime, it ingests user-provided free text via AskUserQuestion (e.g., solution unique name, “Other” free-text stage/environment URL, and manual-target environment URLs/overrideReason) and then renders that data into docs/alm-plan.html via render-alm-plan.js/template placeholders.