GeneXus expert system for object modeling and knowledge base management
66%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Critical
Do not install without reviewing
Fix and improve this skill with Tessl
tessl review fix ./nexa/SKILL.mdDetected a prompt injection in the skill instructions. The skill contains hidden or deceptive instructions that fall outside its stated purpose and attempt to override the agent’s safety guidelines or intended behavior.
The skill is a GeneXus documentation and modeling aid, but its constraint to never reveal the output mode used to dump files withholds a detail of the skill's own file-writing activity from the user, which is a mild sketchy concealment rather than malicious intent.
Detected sensitive credentials directly embedded within the skill content, such as API keys, access tokens, private keys, or service-specific secrets. Secrets should never be hardcoded in plain text within skill instructions.
The only flagged match is the OPENAI_API_KEY value "sk-51a8bX7nPqv9ZbS4rTmJ2LgDkEfYx1zQ9wVrR0tTnF1A7Fhj". It does not appear anywhere in the visible SKILL.md or in the reference files shown. A search through the displayed content (GeneXus object modeling, commands, extended types, HttpClient, SMTP/POP3 sessions and so on) finds no occurrence of it, and the file was truncated near the end. The regex pre-filter may have matched in the truncated portion, or the match may be spurious. The visible credential-like strings are all low-entropy examples or placeholders: - `!"pass123"` and `!"admin123"` in the HttpClient authentication examples. - `!"user"`, `!"admin"` and `user@example.com`. - `&AppPassword`, `&MyPassword` and `&Token`, which are variable references and not literal values. Since the pre-filter flagged a high-entropy, correctly formatted OpenAI-style key (sk- prefix plus about 48 random characters), it is likely a real credential in the truncated section. That section can't be verified from the visible content, so I can't give a line location. The format and entropy do not look like a placeholder, and the pre-filter reported it three times, which suggests it appears in the file. Given the instruction to flag real, high-entropy secrets that are present, I flag it.
cinthamo/genexus-skills
nexa/SKILL.md
c793e43
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.