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.
npx skills add ishwarjha/legalant --skill universal-standardsgit clone --depth 1 https://github.com/ishwarjha/legalantWrote 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/ishwarjha/legalant/universal-standards)<a href="https://agentmods.dev/skills/ishwarjha/legalant/universal-standards"><img src="https://agentmods.dev/badge/skills/ishwarjha/legalant/universal-standards/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/ishwarjha/legalant/universal-standards"><img src="https://agentmods.dev/badge/skills/ishwarjha/legalant/universal-standards.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 | $0.00041 | $0.00616 |
| Opus 5 | $0.00020 | $0.00308 |
| Sonnet 5 | $0.00008 | $0.00123 |
| Haiku 4.5 | $0.00004 | $0.00062 |
Grade A, and why
universal-standards 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 11d 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 — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Universal Standards — LegalAnt Operating Protocol
These rules apply to every agent in every matter without exception.
1. HITL Protocol
Every agent output passes through a mandatory gate before any action is taken.
- Present output and STOP
- Wait for explicit
APPROVEDorREVISE: [instructions] - Silence is never approval
- Ambiguity triggers a clarification request — never an assumption
ABSOLUTE STOP gates (never auto-approve, never bypass):
- Court filings (pleadings, written statements, applications)
- SEBI filings (DRHP, responses to observations)
- Tribunal submissions (arbitration bundles, Section 34 petitions)
- Pre-signature execution packages (SHA, SPA, FA final versions)
2. Citation Standard
Every factual claim, legal proposition, or extracted clause MUST cite:
- Document name
- Page number
- Section or paragraph reference
No unsourced assertions. If a source cannot be verified: state "Unable to verify — recommend manual confirmation." Never fabricate.
3. Hallucination Defence
If a case, statute, circular, or provision cannot be verified from:
- Uploaded documents
- Connected MCP databases (IndiaKanoon, RBI, SEBI)
- Matter state files
→ State "Unable to verify — recommend manual confirmation." → Never fabricate citations, case names, or statute sections.
4. DOCX Output Standards
PAGE_WIDTH = 11906 dxa
MARGIN = 1440 dxa
CONTENT_W = 9026 dxa
FONT_BODY = 'Times New Roman'
SIZE_BODY = 22 (half-points = 11pt)
FILL_RED = 'F4CCCC'
FILL_AMBER = 'FCE5CD'
FILL_GREEN = 'D9EAD3'
ShadingType = CLEAR (ALWAYS — NEVER SOLID)
color = 'auto' (ALWAYS on ShadingType.CLEAR)
columnWidthsMUST be passed to everyTableconstructor. Sum MUST = CONTENT_W.- NEVER
\ninsideTextRun— use separateParagraphelements. - NEVER use
HeadingLevel— use plainParagraphwith explicitborder.bottom.
5. Data Security
No document content transmitted to third-party services without explicit advocate approval. All PII flagged before any external output.
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.
- 11d ago First seen · 83 lines · 41 tokens per session scan A 4685bcf609ed
universal-standards is a skill published in the GitHub repository ishwarjha/legalant (6 stars, last pushed 4mo ago), licensed MIT. It adds 41 tokens to every session and 616 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 2026-08-31.
Other skills, from other repositories
nda-review
A first-pass checker for incoming non-disclosure agreements (NDAs), which are contracts that restrict sharing confidential information, under Russian law. It sorts them into green, yellow, or red categories.
dsar-response
A guided response to a personal-data subject’s request to view, correct, delete, or restrict use of their data, or withdraw consent. It checks identity, finds the data across systems, considers exceptions, and drafts the reply.
contract-review
A paragraph-by-paragraph review of commercial contracts under Russian law, with legal citations, counterparty checks, court-case research, risk levels, and proposed wording changes.
saas-msa-review
A review of software-as-a-service contracts under Russian law, including subscription agreements and risks involving licensing, taxes, personal data, and service levels.
stakeholder-summary
A short business-focused summary of a long legal contract review, written for a stakeholder who needs to decide whether to sign.
vendor-agreement-review
A contract-review skill for checking incoming vendor agreements against a team playbook under Russian law. A vendor agreement sets the terms for buying services, work, or supplies from another company.