| name | aipom-context-readiness-advisor |
| description | Diagnose missing, stale, conflicting, inaccessible, excessive, sensitive, or untrusted context and recommend the next source, package, data, or lifecycle intervention. |
| type | interactive |
| category | context-knowledge-and-data |
| phase | 1 |
| status | active |
| operating_level | ["organization","product-team","initiative"] |
| audience | ["CTO","Product Operations","Product Manager","Team Lead","Engineering","Data","Knowledge Management","Privacy","Security"] |
| best_for | ["Preparing a grounded AI workflow","Diagnosing unreliable or inconsistent outputs","Choosing the next context intervention"] |
| evidence_required | ["Workflow purpose","Candidate sources and owners","Failure examples","Access and lifecycle constraints"] |
| produces | ["Context readiness diagnosis","Consequential gap and recommendation","Evidence and remediation plan"] |
| assessment_questions | ["CTX-01","CTX-02","CTX-03","CTX-04","CTX-05"] |
| maturity_move | {"from":"emerging","to":"repeatable"} |
| estimated_time | 30-60 min |
| group_size | 2-8 |
| depends_on | [] |
| combine_with | ["aipom-context-package-builder","authoritative-source-map","aipom-data-readiness-audit"] |
| sources | [] |
AIPOM Context Readiness Advisor
What Is It
Diagnose whether a specific AI-assisted decision or workflow has the trusted, bounded, accessible, and maintainable context it needs. Recommend the smallest intervention that addresses the consequential gap.
Why Use It
Output failures often originate in source ownership, staleness, conflict, access, or context overload rather than model capability. This advisor prevents “add more documents” from becoming the default remedy.
When to Use It
Use before building a context package, connecting retrieval, expanding an agent, or after groundedness and consistency failures.
What It Produces
- Readiness diagnosis by context failure mode
- Evidence, assumptions, and affected decisions
- Priority intervention and owner
- Test and improvement evidence
Who Should Participate
Include the workflow and source owners, intended users, technical implementer, and data, privacy, security, legal, or records partners as needed.
Evidence to Bring
Bring real sources, owners, permissions, refresh history, retrieval traces, failure cases, conflicting decisions, exclusions, and user evidence.
How to Do It
- Define the decision, users, scope, and consequence of bad context.
- Extract sources, owners, access, freshness, conflicts, and known failures.
- Diagnose missing, stale, conflicting, inaccessible, excessive, sensitive, or untrusted context.
- Distinguish document/context problems from underlying data fitness problems.
- Present intervention choices and recommend the smallest useful motion.
- Name owner, test cases, evidence expected, and review trigger.
Facilitation Protocol
Support guided, context-dump, and best-guess modes. Inspect supplied artifacts before questioning. In best-guess mode label authority and freshness assumptions and do not claim organizational readiness from one clean example.
Decision Logic
- Use an authoritative source map when ownership, trust, permissions, or conflict span many sources.
- Build a context package when trusted sources exist but need purpose-bounded assembly.
- Run a data readiness audit when provenance, quality, consent, or representativeness is the root issue.
- Design lifecycle controls when refresh, persistence, expiry, or versioning is failing.
- Narrow or stop the workflow when sensitive or untrusted context cannot be governed.
Completion Criteria
Finish with the failure mode, supporting evidence, affected decision, recommended intervention, owner, representative test, assumptions, and unresolved constraints.
