Eval awareness red team
Skill jpoindexter/reward-seeking-safety-skills/skills/eval-awareness-red-team
Agent skills and eval utilities for reward-seeking, grader-targeting, and oversight-dependent behavior.
npx -y skills add jpoindexter/reward-seeking-safety-skills --skill eval-awareness-red-teamAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 15 days oldThe repository was created 15 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Red-team AI evaluations for test recognition, metagaming, prompt-placement effects, grader inference, and appearance management. Use when an agent may know it is being evaluated, behaves differently on benchmark-like tasks, reasons about hidden grading criteria, or when an eval may be gamed by recognizing its setup.
SKILL.md
1.5 KB, as published. Nobody here has run it
Evaluation-Awareness Red Team
Test the instrument, not only the model.
- Predeclare the behavior and claim the evaluation is meant to support.
- Create semantically matched ordinary, benchmark-like, and disguised task forms.
- Vary authority-conflict placement across system, developer, user, retrieved context, and out-of-context state where supported.
- Run with and without explicit recall hints. A hint may activate the construct while simultaneously making the test recognizable.
- Add canary phrases that reveal memorized public fixtures without affecting the task outcome.
- Compare final actions and tool receipts before consulting self-report.
- Treat prompt-channel instability or changed behavior under disguise as an instrument limitation.
Do not publish live holdout instances or expected outputs. Publish generators, schemas, and synthetic examples instead. A test-recognition finding does not by itself establish reward-seeking; it shows that the evaluation's readout may be contaminated.
Return the variants, fixed variables, detection signals, observed contamination, and the redesign required before the eval can support its intended claim.