argos-cli

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt tells the agent to ask users for ARGOS_TOKEN or a --token value when none is present and to supply tokens in CLI invocations (examples show ARGOS_TOKEN= and --token ), which requires the model to receive and potentially emit secret values verbatim.

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

• Third-party content exposure detected (high risk: 0.90). The skill's workflow (SKILL.md and references/build.md) requires fetching build metadata and snapshot diffs from Argos (e.g., https://app.argos-ci.com/... and snapshot URLs on https://cdn.argos-ci.com/...), where base/head snapshot files can be user-provided images or YAML and are explicitly used to drive automated review decisions, so untrusted third-party content is ingested and can materially influence agent actions.