Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add sigistry/marketplace/plugin install recordsWrote 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/sigistry/marketplace/fhir-validation)<a href="https://agentmods.dev/skills/sigistry/marketplace/fhir-validation"><img src="https://agentmods.dev/badge/skills/sigistry/marketplace/fhir-validation/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/sigistry/marketplace/fhir-validation"><img src="https://agentmods.dev/badge/skills/sigistry/marketplace/fhir-validation.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 | $0.00103 | $0.01084 |
| Opus 5 | $0.00051 | $0.00542 |
| Sonnet 5 | $0.00021 | $0.00217 |
| Haiku 4.5 | $0.00010 | $0.00108 |
Grade A, and why
fhir-validation 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 6d 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 — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Records FHIR Validation
Validate FHIR resources with Records and local project tooling. Prefer local or already configured execution. Do not send clinical data, patient data, FHIR server URLs, or full patient resources to external services unless the user explicitly consents for this task.
Arguments
Treat the user request as a file, directory, JSON resource, OperationOutcome, FHIR URL, or validation goal. If no target is inferable, ask one concise question.
Bundled Paths
Resolve <skill-root> to the absolute directory containing this SKILL.md. Resolve <plugin-root> two directories above <skill-root>. In Claude Code ${CLAUDE_PLUGIN_ROOT} is also the plugin root; in Codex derive it from the installed skill path. Never resolve bundled scripts relative to the user's workspace.
Fast Start
For file or directory validation, first run the deterministic detector and runtime planner when available:
node "<skill-root>/scripts/detect-fhir-project.mjs" <target-or-repo>
node "<skill-root>/scripts/plan-runtime.mjs" <target-or-repo>
Use their output to pick source directories, generated directories, available runtimes, privacy gates, and validation order.
Then choose the first suitable mode:
- Records MCP tools, when available.
- Records API, only when
RECORDS_API_URLis configured and the privacy boundary is acceptable. - Records CLI, using
records validate-file <target>. - Existing local profile-aware validators such as SUSHI, IG Publisher, Firely Terminal, Java validator, or HAPI when already configured and relevant.
- Structural fallback via
scripts/validate-structural.mjs(scope), clearly labeled as not profile-, terminology-, invariant-, or reference-aware.
Privacy Gates
Ask for explicit consent before fetching FHIR URLs, contacting FHIR servers, sending resources to external APIs, installing validators, or using hosted/non-local validation on data that may contain PHI. Prefer local CLI validation. In summaries, include paths, issue codes, and minimal snippets; do not reproduce complete Patient resources or unnecessary identifiers.
What ships with it
32 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- agents/openai.yaml 425 B
- assets/records-app-icon.svg 1.0 KB
- assets/records-signet.svg 528 B
- references/ci-templates.md 1.7 KB
- references/ig-workflows.md 4.2 KB
- references/operationoutcome-map.md 5.0 KB
- references/quality-rules.md 1.9 KB
- references/repair-policy.md 2.3 KB
- references/structural-validation.md 3.0 KB
- scripts/analyze-structuredefinition.mjs 4.3 KB runs code
- scripts/derive-quality-rules.mjs 3.6 KB runs code
- scripts/detect-fhir-project.mjs 15 KB runs code
- scripts/doctor-packages.mjs 1.8 KB runs code
- scripts/explain-operationoutcome.mjs 1.9 KB runs code
- scripts/generate-ci.mjs 3.6 KB runs code
- scripts/generate-issue-map-doc.mjs 1.9 KB runs code
- scripts/lib/fhirpath-pointer.mjs 4.2 KB runs code
- scripts/lib/operationoutcome-issues.mjs 5.9 KB runs code
- scripts/lib/package-doctor.mjs 3.1 KB runs code
- scripts/lib/process-runner.mjs 1.3 KB runs code
- scripts/lib/r4-primitives.mjs 2.7 KB runs code
- scripts/lib/r4-structural-schema.mjs 7.1 KB runs code
- scripts/lib/result-contract.mjs 1.4 KB runs code
- scripts/lib/runtime-policy.mjs 9.2 KB runs code
- scripts/lib/safe-io.mjs 3.7 KB runs code
- scripts/map-fhir-expression.mjs 839 B runs code
- scripts/map-generated-to-fsh.mjs 4.5 KB runs code
- scripts/match-slices.mjs 6.8 KB runs code
- scripts/plan-runtime.mjs 1.5 KB runs code
- scripts/redact-fhir-summary.mjs 3.8 KB runs code
- scripts/validate-structural.mjs 12 KB runs code
- scripts/validate.mjs 12 KB runs code
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.
- 6d ago First seen · 88 lines · 103 tokens per session scan A 7aac5edc8e96
fhir-validation is a skill published in the GitHub repository sigistry/marketplace (3 stars, last pushed 6d ago), licensed MIT. It adds 103 tokens to every session and 1,084 once invoked, about $0.0005 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 2026-09-03.
Other skills, from other repositories
architecture-optimization
Guided journey from a working codebase grown slow and tangled to one measurably fast, cleanly bounded, and readable. Orchestrates eight skills phase by phase - working-with-legacy-code, clean-architecture, software-design-philosophy, refactoring-patterns, system-design, ddia-systems, release-it, pragmatic-programmer …
improve-code-quality
Guided journey from a working-but-untested vibe-coded prototype to a production-ready product with tests, clean structure, a business-rules boundary, and resilience at scale. Orchestrates nine skills phase by phase - working-with-legacy-code, clean-code, refactoring-patterns, software-design-philosophy…
remove-technical-debt
Guided journey from a large aged codebase everyone fears to touch to one that is safe to change, legible, bounded, and resilient - paid down in place without a rewrite. Orchestrates eight skills phase by phase - working-with-legacy-code, refactoring-patterns, clean-code, software-design-philosophy, clean-architecture…
release-it
Build production-ready systems with stability patterns: circuit breakers, bulkheads, timeouts, and retry logic. Use when the user mentions "production outage", "circuit breaker", "deployment pipeline", "chaos engineering", "retry storm", "health checks", "my service keeps crashing", "prevent cascading failures", or…
ask-codex
Consult OpenAI Codex for investigation, debugging, or code review. Use when user explicitly asks to "ask codex", "check with codex", "codex review", or as a last resort when stuck after 4+ failed attempts at debugging, investigation, or bug fix and completely out of ideas. Codex is slow (2-5 min), so only escalate…
verify
Verify Elixir/Phoenix changes — compile, format, and test in one loop. Use after implementation, before PRs, or after fixing bugs.