critical
suspicious.exposed_secret_literal
- Location
- scripts/twitter_engagement_client.py:304
- Finding
- File appears to expose a hardcoded API secret or token.
Twitter Autopilot
AdvisoryAudited by Static analysis on May 10, 2026.
Overview
Detected: suspicious.exposed_secret_literal
Findings (1)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
What this means
The package may contain an active API key or use a credential the user did not provide, which can create billing, abuse, or account-boundary issues.
Why it was flagged
The skill declares AISA_API_KEY as the credential source, so an apparent literal API key in the bundled script is an undeclared credential and exposes or reuses authority outside the documented boundary.
Skill content
Static scan at line 304: client = TwitterClient(api_key=[REDACTED])
Recommendation
Do not use this release until the hardcoded key is removed, the exposed credential is rotated, and the script uses only the documented AISA_API_KEY configuration.
What this means
If invoked on the wrong content or account, the agent could publish posts or perform likes/follows visible to others.
Why it was flagged
These are public, account-mutating Twitter/X actions. They are disclosed and purpose-aligned, but users should notice that the skill can change public account state once authorized.
Skill content
Also supports posting, liking/unliking tweets, and following/unfollowing users after the user completes OAuth authorization.
Recommendation
Use explicit instructions for each public action, review post content and targets before execution, and revoke OAuth access if you no longer need the skill.
What this means
Any attached image or video selected for posting will leave the local workspace and be sent to the relay/Twitter service.
Why it was flagged
The skill sends selected local media files through the AIsa relay backend to publish them on Twitter/X. This is disclosed and expected for media posting, but it is still an external data transfer.
Skill content
The Python client reads the local file and sends it to the relay backend as multipart/form-data.
Recommendation
Only attach files you intend to publish, avoid private media, and verify the relay URL if using TWITTER_RELAY_BASE_URL.