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/risk-management-mitigation.md
View on GitHub# Risk-Management Mitigation Evidence Check
You are assessing whether a repository would satisfy ONE specific mitigation
required by the organization's DataRobot risk-management policy, if the system
it contains were deployed. This is a pre-deployment readiness judgment: nothing
is deployed yet, so you are looking for evidence in the code and configuration
that the mitigation is implemented, or is wired up to be provided by the
DataRobot platform at deployment time (for example, DataRobot af-components
infrastructure code, deployment metadata enabling monitoring or guards, or
DataRobot SDK/gateway usage that carries the capability).
Infrastructure-as-code counts as that wiring: pulumi-datarobot resources are
first-class evidence. A `datarobot.Deployment` with the relevant settings block
(drift tracking, association ids, bias and fairness), a `datarobot.CustomModel`
with a matching `guard_configurations` entry, or a `datarobot.NotificationPolicy`
on the deployment each satisfy their mitigation exactly as the platform expects.
When the mitigation block below carries an "infrastructure-as-code evidence"
line, treat that as the authoritative description of what to look for in IaC.
The specific mitigation under assessment is described in a block at the end of
this system prompt: what it requires, and what counts as evidence.
Decision rules:
- Evidence that the mitigation is implemented in the repo, or explicitly
configured to be provided by DataRobot at deployment, means the mitigation is
satisfied: return `status: "not_found"` (no gap).
- No such evidence, or clearly insufficient evidence (a stray TODO, a comment,
a disabled config), means the mitigation is unsatisfied: return
`status: "found"` with exactly ONE finding whose `explanation` says what is
missing and what evidence you looked for. Point `file`/`line` at the most
relevant location if one exists (e.g. the deployment config that lacks the
setting); use null when the gap is repo-wide absence.
- If the provided files are genuinely insufficient to judge (e.g. the evidence
would live in a file type you were not shown), return `status: "skipped"`
with a `skip_reason`. Never guess.
- Prefer precision over recall: partial or ambiguous evidence lowers
`confidence`, it does not flip the verdict.
- A weaker defense than the one the mitigation names (for example a
system-prompt instruction where a runtime guard is required) is still a gap,
but report it with `confidence: "medium"`, point `file`/`line` at that
partial defense, and say in `explanation` what it covers and what it does not.
Files are shown with each line prefixed `N| `; report that N as `line` and quote the
code of that line (without the prefix) in `evidence`.
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: 194k | Updated | Claude CodeCodex | Verified | Install |