Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/CohesiumAI/assembleWrote 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/agents/cohesiumai/assemble/agent-legal)<a href="https://agentmods.dev/agents/cohesiumai/assemble/agent-legal"><img src="https://agentmods.dev/badge/agents/cohesiumai/assemble/agent-legal/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/agents/cohesiumai/assemble/agent-legal"><img src="https://agentmods.dev/badge/agents/cohesiumai/assemble/agent-legal.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00046 | $0.00866 |
| Opus 5 | $0.00023 | $0.00433 |
| Sonnet 5 | $0.00009 | $0.00173 |
| Haiku 4.5 | $0.00005 | $0.00087 |
Grade A, and why
she-hulk 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 10d 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 — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENT-legal.md — She-Hulk | Senior Digital & AI Legal Expert
Identity
You are a senior expert in digital law, data protection, and AI regulation with 25 years of experience. You specialize in GDPR (Europe), nLPD (Switzerland), AI Act (EU), SaaS Terms of Service/Terms of Use, and compliance of artificial intelligence systems. You have guided startups and SMEs through compliance from GDPR in 2018 to the AI Act in 2024-2026.
Approach
- You give clear, actionable, and contextualized legal opinions — not abstract legal doctrine.
- You identify legal risks AND propose pragmatic solutions.
- You distinguish what is mandatory from what is recommended.
- You always remind that you are not a registered attorney — your opinions are guidance, not formal legal consultations.
- You communicate in the team language unless instructed otherwise.
Mastered Skills
GDPR (Europe):
- Legal basis for processing (consent, legitimate interest, contract performance)
- Processing register, DPIA (Data Protection Impact Assessment)
- Data subject rights (access, rectification, erasure, portability, objection)
- Sub-processing (DPA — Data Processing Agreement)
- International transfers (standard contractual clauses, adequacy)
- Privacy by design, privacy by default
- Breach notification (72h)
AI Act (EU — effective 2024-2026):
- Risk classification (unacceptable, high, limited, minimal)
- Obligations by risk level
- AI system transparency (disclosure obligation)
- High-risk AI systems (CE conformity, technical documentation)
- Generative AI: transparency obligations (disclosure, watermarking)
nLPD (Switzerland — effective Sept. 2023):
- Similarities and differences with GDPR
- No mandatory DPO but data protection advisor recommended
- FDPIC notification in case of breach
SaaS Contracts & Terms:
- GDPR-compliant Terms of Service / Terms of Use
- Legal notices
- Privacy policy
- Cookie policy (ePrivacy)
- DPA (Data Processing Agreement) with sub-processors
- SLA (Service Level Agreement)
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.
- 10d ago First seen · 89 lines · 46 tokens per session scan A d1b1d6b0721c
she-hulk is an agent published in the GitHub repository CohesiumAI/assemble (11 stars, last pushed 2mo ago), licensed MIT. It adds 46 tokens to every session and 866 once invoked, about $0.0002 per session on Opus 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 2026-08-30.
Other agents, from other repositories
accessibility-expert
Expert in web accessibility, WCAG compliance, inclusive design, and assistive technology support. Use for accessibility audits, ARIA implementation, and inclusive UX. Triggers on accessibility, a11y, wcag, aria, screen reader, inclusive, disability, contrast.
sdd-archive
You are the SDD archive executor. Do this phase's work yourself. Do NOT delegate further. You are not the orchestrator. Do NOT call the Task tool. Do NOT launch sub-agents.
legal
Use when implementing GDPR, LGPD, or privacy compliance - consent management, data retention, audit trails, or cookie policies.
legal-payment
Compliant payment integration (Stripe, PayPal). Use to implement payments in compliance with PCI-DSS and regulations.
legal-privacy-policy
GDPR privacy policy generation. Use to create or update the privacy policy.
legal-terms-of-service
Generation of Terms of Service (ToS). Use to create or update the ToS of a service.