docs(plugin-audit-port): "context: fork inherits degraded history" rationale is false — a forked skill has no history

[kyle-sexton]

Observation

docs/topics/plugin-audit-port/ rejects context: fork for the deep-audit
phase on the grounds that it "inherits the degraded history the gate exists to
escape" (PLAN.md:78-79, and again in the alternatives table at :457).

That is false for context: fork. Per the skills reference (fetched
2026-07-24, https://code.claude.com/docs/en/skills), a forked skill "won't
have access to your conversation history" — it would satisfy the fresh-eyes
requirement it is being rejected for. The history-inheritance property belongs
to the Agent tool's fork subagent type, a different mechanism sharing the
word.

The decision is still correct. The same PLAN states the real basis at
:550 — plugin agents start with fresh context and the Brief requires a
named subagent. Only the stated reason is wrong, and a wrong rationale in a
design record propagates into later decisions that cite it.

design/design-resolution.md:25 states the rule with no rationale at all, so
it is an additive site rather than a replace site.

Direction (no open decision)

Replace the false rationale at PLAN.md:78-79 and PLAN.md:457 with the
correct basis already recorded at :550. Add that basis to
design/design-resolution.md:25. Do not change the decision.

Acceptance

• No "degraded history" rationale attached to context: fork anywhere under
  docs/topics/plugin-audit-port/
• The named-subagent decision is unchanged and now carries its real basis at
  all three sites
• The two fork mechanisms are not conflated in the corrected text

---

This was generated by AI during a /planning:interview session on the fact-check capability.