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/vaquarkhan/compliance-agent-skillsWrote 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/rules/vaquarkhan/compliance-agent-skills/20-phi-redaction-first)<a href="https://agentmods.dev/rules/vaquarkhan/compliance-agent-skills/20-phi-redaction-first"><img src="https://agentmods.dev/badge/rules/vaquarkhan/compliance-agent-skills/20-phi-redaction-first/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/rules/vaquarkhan/compliance-agent-skills/20-phi-redaction-first"><img src="https://agentmods.dev/badge/rules/vaquarkhan/compliance-agent-skills/20-phi-redaction-first.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 | $0.00285 | $0.00285 |
| Opus 5 | $0.00143 | $0.00143 |
| Sonnet 5 | $0.00057 | $0.00057 |
| Haiku 4.5 | $0.00028 | $0.00028 |
Grade A, and why
20-phi-redaction-first 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 9d 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.
What it actually says
PHI Redaction First
Upstream gate
redaction.py + Presidio analyze all user prompts before agent.py sends text to the model.
Token handling
- Masked tokens like
<PERSON_1>are intentional — do not guess original values. - Deanonymization is restricted to authorized downstream delivery via
deanonymize_responsetool. - Between engagements, assume a fresh redaction session (
reset_session()).
Entity coverage
Default types include PERSON, US_SSN, PHONE_NUMBER, EMAIL_ADDRESS, MEDICAL_LICENSE, DATE_TIME, CREDIT_CARD, and others — see DEFAULT_ENTITY_TYPES in redaction.py.
Hooks
COMPLIANCE_PHI_REDACTION=required — session hooks block work if Presidio unavailable.
Skill
For pipeline tuning: hipaa-phi-redaction-pipeline
Reference: references/phi-redaction-patterns.md
PCI note
Credit card entity detection helps prevent accidental PAN in prompts — never process PAN in LLM workflows regardless of redaction.
Minimum necessary
Scope LLM access to redacted, minimum-necessary text. Prefer structured queries over full clinical notes.
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.
- 9d ago First seen · 39 lines · 285 tokens per session scan A f761067e68f3
20-phi-redaction-first is a cursor rule published in the GitHub repository vaquarkhan/compliance-agent-skills (2 stars, last pushed 16d ago), licensed MIT. It adds 285 tokens to every session, about $0.0014 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 cursor rules, from other repositories
compliance-trestle
Compliance Trestle repository rules.
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
cli-error-handling
CLI command error handling patterns.
prefer-assertions-over-defensive-checks
Prefer assertions over defensive checks when data is guaranteed to be valid.