Blog Draft

This content cannot be published — it is an internal production flag, not a blog post.

The story dossier for this piece rates its own evidence at 2/5 and contains an explicit stop-work directive. Three critical evidence gaps block drafting: alternatives considered (speculative, no interview conducted), execution details (almost entirely missing), and measurable outcomes (entirely absent). Publishing invented practitioner reasoning as organizational experience is an auto-fail condition under the evidence gate this content system operates on.

If you are seeing this in a CMS preview, this entry should be unpublished immediately and routed back to the editorial queue. The steps required before this piece can be drafted are documented in the production flag below.

Why This Piece Is Blocked

The evidence gate for Tier 1 practitioner blogs requires recoverable organizational experience — not inference. The team built the artifact. The reasoning behind how and why they built it does not yet exist in any source available for drafting. Three gaps make the piece undraftable:

• Alternatives considered: Marked as weak and speculative. No engineering-lead interview has been conducted. Any rejection rationale presented in a draft would be fabricated.
• Execution details: Redaction logic, versioning strategy, MCP protocol selection rationale, schema design, team composition, and development timeline are all unknown.
• Measurable outcomes: No metrics, no qualitative feedback, and no before/after comparison exist in the dossier.

Additionally, a duplication risk against Story #1 — already published, covering the same engineering artifact, the same AI coaching tool, and the same core tension — has not been resolved.

What Must Happen Before Drafting Can Begin

Step 1 — Duplication check (immediate, approximately 15 minutes)

Review the published Story #1 blog and map its coverage against this story's proposed angles. If MCP architecture, redaction logic, and the privacy-boundary framing are substantially covered, Story #2 should be killed, merged as an update, or reframed to a genuinely distinct angle before any drafting work begins.

Step 2 — Commission two practitioner interviews (if duplication check clears)

Interview 1: Engineering lead on /mcp/forta

• What alternatives did the team actually evaluate and reject, and why?
• Why MCP over a custom API or direct integration?
• How does the redaction logic work — what gets redacted, by what rules, and where in the pipeline?
• What is the versioning strategy, and what triggers a new version?
• What was the team composition and development timeline?
• What broke or surprised the team during implementation?

Interview 2: Coach and Growth product owner

• What did the grounding gap actually cost in coaching quality and practitioner trust before the MCP server shipped?
• What changed after it shipped — any qualitative signals, even anecdotal?
• What adoption patterns shifted, if any?

Step 3 — Return the enriched dossier for drafting

With interview material in hand, the dossier's evidence grade needs to reach at least 4/5 before a Tier 1 blog can be responsibly produced. Interviews must recover alternatives-considered reasoning (mandatory), execution detail (mandatory), and at least one outcome signal — quantitative or qualitative (strongly preferred).

Recommended Next Step for the Production Queue

Advance immediately to Tier 1, Story #3 — "How We Recovered a Dual-Brand Platform Merge Mid-Flight and Still Hit Go-Live." That story carries no duplication risk and is independently strong as a Tier 1 candidate. Story #2 returns to the queue once the duplication check resolves and at least one practitioner interview is completed. If interviews yield thin results, the story downgrades to a LinkedIn post or merges into Story #1.

This is a stop-work flag routed through the content system, not a publishable post. Do not publish. Route to editorial for evidence gathering per the evidence gate in E.0.