OpenEnv is a framework for creating, deploying, and using isolated execution environments through simple APIs modeled on Gymnasium. It is intended for researchers and developers training agents with reinforcement learning, as well as creators who need secure environments that can be deployed and accessed through familiar protocols. The catalogue add-ons provide agent integrations and workflows for using OpenEnv.
Borrowing it
Nothing to install: this file belongs to huggingface/OpenEnv. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/huggingface/OpenEnv/main/.claude/skills/alignment-review/SKILL.mdgit clone --depth 1 https://github.com/huggingface/OpenEnvWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/huggingface/openenv/alignment-review)<a href="https://agentmods.dev/skills/huggingface/openenv/alignment-review"><img src="https://agentmods.dev/badge/skills/huggingface/openenv/alignment-review/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/huggingface/openenv/alignment-review"><img src="https://agentmods.dev/badge/skills/huggingface/openenv/alignment-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00042 | $0.00788 |
| Opus 5.5 | $0.00017 | $0.00315 |
| Sonnet 5.5 | $0.00008 | $0.00158 |
| Haiku 4.5 | $0.00004 | $0.00079 |
Grade A, and why
alignment-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Alignment Review
Review code changes for alignment with OpenEnv principles using a two-tier model.
Instructions
-
Run automated checks first:
- Execute
bash .claude/hooks/lint.sh- capture lint issues - Execute
bash .claude/hooks/check-debug.sh- capture debug code
- Execute
-
Read alignment documents:
.claude/docs/PRINCIPLES.md- design principles.claude/docs/INVARIANTS.md- system invariants
-
Read open RFCs:
- Scan
rfcs/directory for all RFC files - Note the status of each RFC (Draft, In Review, Accepted, Implemented)
- Pay special attention to Draft and In Review RFCs - these represent active design discussions
- Scan
-
Analyze changes (use
git diffor provided diff):- Identify mechanical issues (Tier 1)
- Flag alignment concerns (Tier 2)
- Flag conflicts with open RFCs (Tier 2)
Tier 1: Uncontentious Issues (Fix Immediately)
These are issues to fix without human input:
- Lint failures from hook output
- Debug code from hook output (print statements, breakpoints)
- Uninitialized variables, type errors
- Missing imports, syntax errors
- Security issues (credential exposure, injection vulnerabilities)
Tier 2: Alignment Discussion Points
For each potential alignment concern, format as:
**ALIGNMENT FLAG**: [Brief description]
- **Principle/RFC at stake**: [Which principle from PRINCIPLES.md or RFC number]
- **The concern**: [What seems misaligned or in conflict]
- **Suggested reviewer**: @darktex [pull actual reviewers based on authors of the specific line of PRINCIPLES.md and INVARIANTS.md using git blame, and/or authors of conflicting RFCs]
Examples of Tier 2 Issues
Principle conflicts:
- Adding external reward computation (violates "rewards in environment")
- Client importing server code (violates client-server separation)
- New API that differs from Gymnasium pattern
RFC conflicts (flag even for Draft/In Review RFCs):
- Change conflicts with design proposed in an open RFC
- Change pre-empts a decision being discussed in an RFC
- Change implements something differently than an RFC proposes
- Change affects an area covered by an RFC under review
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 95 lines · 42 tokens per session scan A 511d55d37baa
alignment-review is a skill published in the GitHub repository huggingface/OpenEnv (2,739 stars, last pushed 8d ago), licensed BSD-3-Clause. It adds 42 tokens to every session and 788 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
autoreview
Structured code review through the shared OpenClaw agent-skills installation.
omh-code-review
This is an OMH code-review workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
explore-unknowns
Guide the user through a quadrant walk that maps the unknowns of a task — open by listing the known knowns, then work through known unknowns, unknown knowns, and unknown unknowns one stage at a time, ending with a complete four-quadrant map in the user's hands. Use when a request is ambiguous or underspecified, the…
re0-merge
Review and land an external contribution the way this suite does: gate it against the thesis, land it with the author's credit intact, complete a new skill rather than merging it raw, then approve, credit, and explain before closing. Use when reviewing a pull request, as any collaborator or maintainer, not only the…
asplos-writing-style
Use when drafting or revising ASPLOS prose — engineering the first two pages that rapid review actually reads, stating the cross-layer insight as one quotable sentence, structuring per-layer mechanism sections, writing bounded claims, and fitting everything into 11 self-contained pages that include figures and…