arman-jalili/Rigorix

EU AI Act compliance & audit — AI code governance with policy enforcement, audit trails, code generation, self-correcting validation

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
modeExecution mode: - auto: Detect from event context (PR opened → Mode A, issue_comment → Mode B) - governance: Mode A — policy checks on PR diffs - run: Mode B — full plan → execute → persist - validate: Mode B — run with self-correcting validation loop - plan: Mode B — plan only, no execution - status: Show current execution status noauto
intentNatural-language intent for Mode B (run/validate/plan). What should Rigorix do?no
permission-modePermission mode for engine tool execution: - read_only: Only read operations - workspace_write: Read + write within workspace (CI default) - dangerous_full_access: No restrictions noworkspace_write
policy-filePath to policy.toml for Mode A governance. Loaded from BASE BRANCH to prevent tampering.no.rigorix/policy.toml
fail-on-violationMode A: Fail workflow on policy violations. Default false (warn-only for early adoption).nofalse
fail-on-action-errorFail workflow if the action itself encounters an error. Default false (fail-open).nofalse
max-llm-callsMaximum LLM API calls per executionno50
max-llm-tokensMaximum LLM tokens per executionno50000
max-validation-iterationsMaximum validation loop iterationsno3
max-retriesMaximum retries for transient API failuresno3
retry-delay-msBase retry delay in milliseconds (exponential backoff with ±25% jitter)no1000
post-pr-commentPost execution summary as a PR commentnotrue
profileConfiguration profile to useno
providerLLM provider (anthropic, openai, deepseek, custom)noanthropic
api-base-urlLLM API base URL (e.g. https://api.deepseek.com/v1 for DeepSeek, https://api.openai.com/v1 for OpenAI)no
modelLLM model name (e.g. deepseek-chat, gpt-4o, claude-sonnet-4-20250514)no
max-tokensMaximum tokens per LLM callno4096
temperatureLLM temperature (0.0-1.0)no0.3
backend-api-keyAPI key for the audit/enforcement backendno
backend-audit-urlURL for posting audit records (e.g. https://backend.example.com/audit)no
namedescription
execution_idUUID of the execution
statusFinal execution status: completed, failed, partial_failure
mode_usedWhich mode was used: governance, run, validate, plan
iterationsNumber of validation loop iterations
template_idID of the generated or used template
quality_levelAchieved quality level: targeted_tests, package, workspace, merge_ready
failure_countNumber of failures (0 on success)
cumulative_tokensTotal LLM tokens consumed
duration_msTotal execution duration in milliseconds
violations_countMode A: Number of policy violations detected