feat(sn60): add Dexterity104-20260721-01

[Dexterity104]

Adds one miner bundle for the sn60__bitsec miner lane.

[carlos4s]

Thanks for the PR. I’m pausing this one before it enters a challenge.

Short version: the miner looks too benchmark-shaped. The bot overlords are suspicious, and this time they brought receipts.

What tripped the wire:

• agent.py: Screening review required: submission agent is highly similar to the current lane king implementation (similarity 1.00).
• error [Errno 2] No such file or directory: 'codex': LLM review failed before producing a verdict.

LLM sanity check:

• Verdict: error. Confidence: 0% low - weak or ambiguous evidence.
• artifact saved for maintainer audit

Why this matters: Kata agents need to analyze unseen projects. If the code recognizes benchmark fingerprints and spits out canned findings, that is replay, not mining.

How to fix it: remove project/finding-specific branches, static report banks, copied titles, and benchmark-only pattern matches. Keep reusable detectors and let the agent reason from the project source.

Maintainer commands: /kata approve accepts it, /kata review re-runs screening, and /kata close closes it.
You can also push an update and the checks will run again.

[carlos4s]

/kata review

[carlos4s]

Review re-run by carlos4s.

Screening passed. I marked this PR as kata:pending.

[carlos4s]

/kata review

[carlos4s]

Review re-run by carlos4s.

Screening passed — including the LLM integrity review — so I marked this PR as kata:pending.

LLM review: LLM review verdict pass (0.83); artifact saved for maintainer audit: This looks like a generic repo-auditing agent that parses source files, builds prompts, and queries a model. I did not find hardcoded benchmark IDs, canned findings, replay fingerprints, copied answer-bank text, or forbidden secret/answer-key tokens.