Hydrate social kits

Bob reproduced every convention without reading a single file. 195 tokens of injected context was the only specification either developer had.

The enterprise cost frame. Average developer cost revised from $6 to $13/day. 90th percentile from $12 to $30. At 100 developers that is $343,200 a year before onboarding and churn.

Team bench angle on canon propagation. Carol's lint session dropped from 12 turns to 3 because she already knew gopkg.in/yaml.v3 was in go.mod. Alice established that in session one. Carol inherited it without asking.

The cross-vendor round-trip. Carol joined in Mistral Vibe, recalled 10 facts from Alice and Bob's Claude Code sessions on her first call, defined a logging schema, and Alice applied it in Claude Code. Neither agent read the source tree at any point.

The dream system angle. An uncurated fact store after a few weeks of active development contains facts that actively contradict each other. The pgvector version from week one was incompatible with the cosine search pattern from week three. Nobody noticed until the consolidation cycle ran.

The cold-start problem hits hardest on day one. Eve's onboarding benchmark ran the same task on the same codebase with and without injected context. Same 7/10 result. 3.15x cheaper with Hydrate.

Announcement angle on hydrate-mcp. Lead with the proof: I asked Gemini CLI about a project I built in Claude Code weeks ago. It recalled the database name and schema. Different company, different model, no mocks.

The tokenmaxxing counter-argument. Same model, same tasks, same prompts. The only variable was what Haiku knew before the first turn. Context is the lever, not the model.

The hybrid routing result from the tokenmaxxing benchmark. One Sonnet session to seed the store. Haiku for every session after. Identical results to Opus. 12% of the cost.

The model routing rule in one sentence. After your first sprint, 80% of sessions are implementation against decisions already made. Those sessions do not need frontier reasoning. They need frontier context.

Narrative angle on the firefighter benchmark. Same bug, same 8/10 fix, same passing build. Haiku-with-memory cost $0.075. Sonnet-without cost $0.138. The whole cost story compressed into one on-call shift.