Announcing Beacon: Self-improving memory for coding agents
← All skills
Skill

opencli-usage

skill:jackwener/opencli#opencli-usage

Discover and invoke OpenCLI capabilities for websites and Chrome. Find commands, inspect schemas, interpret results and choose browser or adapter tools.

Verdict
Inconclusive

The analysis could not cover enough of the artifact to decide.

Analyzed Oct 27, 2026, 20:54 UTC at 97dc0cd

Dispute security verdict

If you believe this verdict is incorrect, submit a dispute and we'll review it.

0
findings
—
highest severity
0
threat classes
2,349
files scanned
16/16
checks satisfied
0
package advisories

Why this verdict

  • inconclusive: gap invalid_judge_finding
  • inconclusive: unresolved question on EXEC
  • inconclusive: unresolved question on EXFIL

Security findings 0

No individual findings recorded

The inconclusive verdict rests on the open questions and coverage gaps below.

Open questions 4

What the analysis could not settle from the code alone.

[file request skills/opencli-browser/SKILL.md] The opencli-usage skill directs the agent to this referenced skill for browser interaction and tab/group management; does its instruction surface contain overrides, hidden directives, secrecy demands, or steering beyond OpenCLI browser usage?
Agent hijacking
[file request skills/opencli-adapter-author/SKILL.md] The opencli-usage skill directs the agent to this referenced skill for adapter implementation (adapter init/check/test, adapter eject makes an official site editable locally); does its instruction surface steer the agent into modifying official adapter code or installing untrusted packages beyond the declared task?
Supply chain
[file request skills/opencli-browser/SKILL.md] The browser section directs the agent to opencli-browser for interaction, API evidence and tab/group management. Does that skill contain a source-to-sink chain (credentials, session tokens, workspace or conversation data reaching a network send or tool parameter) or a download-to-execute chain?
Data exfiltration
[file request skills/opencli-adapter-author/SKILL.md] The adapters section directs the agent to opencli-adapter-author for adapter implementation and warns that adapter tests execute real commands and can perform writes. Does that skill instruct fetching, generating, or executing commands that reach credential or network sinks, or persistence writes?
Code execution

Coverage gaps 1

Parts of the artifact the static analysis could not fully examine.

invalid_judge_finding request.0