Package index • llmshieldr

Policies and Rules

build_policy(): Build a guardrail policy
policy(): Load a built-in policy
available_policies(): List available built-in policies
policy_controls(): Configure policy-level control actions
add_rule(): Add a rule to a policy
remove_rule(): Remove a rule from a policy
list_rules(): List policy rules
rule_injection_basic() rule_injection_indirect() rule_nlp_intent() rule_pii_email() rule_pii_phone() rule_pii_ssn() rule_secrets_api_key() rule_secrets_bearer() rule_secrets_aws() rule_secrets_password() rule_phi_condition() rule_agency_language() rule_system_prompt_leak() rule_diagnosis_claim() rule_financial_advice(): Built-in rule helpers

Scanners

scan_prompt(): Scan a prompt
preflight_check(): Preflight-check a prompt
scan_context(): Scan retrieved context rows
scan_output(): Scan model output
scan_conversation(): Scan a conversation while preserving message roles
scan_stream(): Scan streamed output chunks with rolling context
stream_guard(): Create a stateful safe-release stream guard
scan_tool_call(): Scan a tool call before execution
scan_tool_output(): Scan tool output before it re-enters model context
scan_url_target(): Check a URL immediately before network use
scan_grounding(): Check output citations against admitted sources
scan_document(): Scan extracted document content
scanner_options(): Configure optional text scanners
redaction_strategy(): Configure redaction behavior
output_contract(): Define a destination-specific output contract
validate_output_contract(): Validate output against a contract
context_policy(): Define retrieved-context admission requirements
grounding_policy(): Configure citation and grounding checks
url_policy(): Configure URL destination policy
document_input(): Describe extracted document content
entity_recognizer(): Create a custom entity recognizer
native_recognizers(): Native checksum-aware recognizers
secret_registry(): Configure versioned secret detection

Orchestration

secure_chat(): Run a guarded chat call
shield_ollama(): Guard an Ollama chat workflow (deprecated)
shield_gemini(): Guard a Gemini Developer API chat workflow (deprecated)
ollama_reviewer(): Create a local Ollama reviewer
remote_reviewer(): Create a remote semantic reviewer
trust_boundary(): Wrap a chat object in a trust boundary
rate_guard(): Create or check a rate guard
tool_policy(): Define authorization and limits for model tools
guard_tool(): Dispatch one tool through guardrails
telemetry_options(): Configure privacy-safe guardrail telemetry

Optional Detector Adapters

guardrail_provider(): Create an optional guardrail provider
prompt_guard_provider(): Adapt a prompt-injection classifier
presidio_provider(): Optional Presidio detector adapter
gitleaks_provider(): Optional Gitleaks secret detector adapter
opa_provider(): Optional Open Policy Agent adapter

Audit and Explanation

write_audit_log(): Write an audit log
explain_findings(): Explain findings
example_prompts(): Example prompts
evaluate_security_cases(): Evaluate scanner behavior on a labeled corpus
summarize_security_evaluation(): Summarize a labeled security evaluation
compare_policies(): Compare policy decisions on the same corpus
owasp_crosswalk(): OWASP LLM Top 10 2026 migration crosswalk
reviewer_prompt(): Return the default semantic reviewer prompt

Constructors

shieldr_rule(): Construct a shieldr_rule
shieldr_policy(): Construct a shieldr_policy
shieldr_report(): Construct a shieldr_report
shieldr_audit(): Construct a shieldr_audit
shieldr_result(): Construct a shieldr_result
print(<shieldr_policy>): Print a shieldr_policy
print(<shieldr_report>): Print a shieldr_report