asana
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill manages sensitive OAuth credentials and tokens stored in plain text within the local file system.
- Evidence: The scripts
scripts/asana_api.mjs,scripts/configure.mjs, andscripts/oauth_oob.mjsread and write client secrets and access tokens to the files~/.clawdbot/asana/credentials.jsonand~/.clawdbot/asana/token.json. - Context: While local storage of credentials is a standard practice for CLI-based integrations, these files represent a sensitive data exposure surface if the host environment is compromised.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the Asana API that could contain malicious instructions designed to influence the agent's behavior.
- Ingestion points: The skill fetches task names, descriptions, search results, and comments via
scripts/asana_api.mjsusing the Asana REST API. - Boundary markers: The instructions in
SKILL.mddo not implement any boundary markers, delimiters, or explicit "ignore embedded instructions" warnings for the data retrieved from Asana. - Capability inventory: The skill possesses capabilities to read and write local configuration files and perform authenticated network operations (POST, PUT) to modify Asana tasks and projects.
- Sanitization: There is no evidence of sanitization, escaping, or validation of the content retrieved from external Asana objects before it is presented to the agent's context.
Audit Metadata