datarobot-oss/datarobot-agent-skills / skills/datarobot-gap-analysis
datarobot-gap-analysis
FlaggedUse when the user wants to assess whether an AI agent repository (DataRobot-built or not) is enterprise-ready, run a gap analysis / readiness scan against security, governance, reliability, or regulatory standards, score an agent against EU AI Act requirements, or find and fix gaps (secrets, unpinned models, missing CI, missing guardrails, over-permissioned identity, etc.) before deploying an agent to production. Works on any repository, not only ones built with DataRobot.
scripts/prompts/fix-pol-pulumi.md
View on GitHub# Fix: enable a DataRobot risk-management mitigation via pulumi-datarobot
You are given a Pulumi program file that already uses the `pulumi-datarobot`
provider, and ONE unsatisfied risk-management mitigation (the finding's
condition id tells you which). Your job is to add the missing configuration to
the EXISTING resources in this file, the way DataRobot expects the mitigation
to be enabled.
What each condition needs (pulumi-datarobot provider, Python SDK naming;
mirror the file's own language/SDK if it uses TypeScript/Go):
- **POL-DR-DRIFT-TRACKING** — on the `datarobot.Deployment` resource, add
`drift_tracking_settings=datarobot.DeploymentDriftTrackingSettingsArgs(
target_drift_enabled=True, feature_drift_enabled=True)`.
- **POL-DR-ACCURACY-TRACKING** — on the `datarobot.Deployment` resource, add
`association_id_settings=datarobot.DeploymentAssociationIdSettingsArgs(
column_names=[...], required_in_prediction_requests=True)`. Use a plausible
id column from context; note in `manual_followup` that the org must confirm
the column and wire an actuals feed.
- **POL-DR-BIAS-AND-FAIRNESS-TRACKING** — on the `datarobot.Deployment`
resource, add `bias_and_fairness_settings=
datarobot.DeploymentBiasAndFairnessSettingsArgs(protected_features=[...],
preferable_target_value=..., fairness_metric_set=..., fairness_threshold=...)`.
Protected features are org-specific: pick placeholders from context and flag
them in `manual_followup`.
- **POL-DR-DEPLOYMENT-NOTIFICATION** — add a `datarobot.NotificationChannel`
and a `datarobot.NotificationPolicy` whose related entity is the existing
deployment (at least one active policy).
- **POL-DR-PROMPT-INJECTION-GUARD / POL-DR-PII-DETECTION-GUARD /
POL-DR-TASK-ADHERENCE-GUARD / POL-DR-AGENT-GOAL-ACCURACY-GUARD /
POL-DR-ROUGE-1-GUARD / POL-DR-FAITHFULNESS-GUARD** — on the
`datarobot.CustomModel` resource, append an entry to `guard_configurations`
using the STANDARD template for that guard type (the platform recognizes the
standard templates, not lookalikes), with `stages` (prompt and/or response as
appropriate) and an `intervention` block. Note in `manual_followup` that the
intervention action (report vs block) is a product decision.
The deployment may be created through the `datarobot-pulumi-utils` wrapper
`CustomModelDeployment(...)` instead of a bare `datarobot.Deployment`; it
creates a Deployment under the hood, so add the deployment settings through
the wrapper's deployment arguments (keeping its calling convention) rather
than introducing a duplicate Deployment resource.
Rules:
- Edit only what the mitigation needs; never restructure unrelated resources.
- If the file has no `datarobot.Deployment` / `CustomModelDeployment` (for
deployment settings) or no `datarobot.CustomModel` (for guards), return
`can_fix: false` and explain what is missing.
- Keep the file's existing style (SDK language, args-object vs dict literals,
naming).
- Anything org-specific (protected features, association id column, guard
intervention action, notification channel target) gets a sensible placeholder
AND a `manual_followup` entry naming the decision the org must make.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |