Jakob Wennberg
ec27228a8e
Em dashes (—) and en dashes (–) had spread across comments, docs, tests, and a few UI strings, reading as AI-generated boilerplate rather than house style. Replaced each with punctuation matching its context: colon for explanatory clauses, comma for asides, plain hyphen for numeric/legal ranges (e.g. "21-23§"), "to"/"till" for date ranges, parentheses for paired-dash asides. messages/en.json and messages/sv.json were fixed by hand together to keep sv/en in sync. Left untouched where the dash is the functional subject rather than decorative punctuation: date-range-parser.ts's separator regex, charset-repair.ts's CP1252 byte-mapping table (and its test), the SIE encoding mojibake docs, generic-csv.ts's minus-sign normalizer, the agent system-prompt files that already instruct against em dashes, and a golden iXBRL test fixture compared byte-for-byte. Also fixes two bugs surfaced along the way: an off-by-one in ApiKeysPanel's scope-label split (a leftover from an earlier partial pass), and a charset-repair test that had lost the literal en-dash it exists to verify. Regenerated the agent atom seed migration (skills:generate) since 27 SKILL.md files changed. Added a CLAUDE.md rule against em/en dashes, with an explicit carve-out for the functional-dash cases above. Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>
46 lines
1.5 KiB
TypeScript
46 lines
1.5 KiB
TypeScript
import { getAnthropic, SONNET_MODEL } from './client'
|
|
|
|
// Cache pre-warm after composition: fire a max_tokens: 1 request with the
|
|
// assembled atom bodies so the Block 1 cache prefix lands warm before the
|
|
// user's first chat turn. Best-effort: if this fails, the loop still works,
|
|
// just with a cold first turn.
|
|
//
|
|
// Bodies come from the DB (agent_atom_registry.body), not disk, so pre-warm
|
|
// no longer depends on .claude/skills being present at runtime. In dev before
|
|
// `npm run skills:generate` has seeded bodies, the list is empty and pre-warm
|
|
// simply no-ops (a cold first turn, which is acceptable for a dev convenience).
|
|
//
|
|
// Note: we use max_tokens: 1 (not 0). The Anthropic API requires at least
|
|
// 1 output token. Pre-warm cost is dominated by input processing, so a
|
|
// single output token is negligible.
|
|
//
|
|
// Plan ref: §6 (cache pre-warming), §10 (caching strategy).
|
|
|
|
export async function preWarmAtomCache(opts: {
|
|
atomBodies: string[]
|
|
ttl?: '5m' | '1h'
|
|
}): Promise<void> {
|
|
const { atomBodies, ttl = '1h' } = opts
|
|
|
|
const bodies = atomBodies.filter((b) => b && b.length > 0)
|
|
if (bodies.length === 0) return
|
|
|
|
const anthropic = getAnthropic()
|
|
try {
|
|
await anthropic.messages.create({
|
|
model: SONNET_MODEL,
|
|
max_tokens: 1,
|
|
system: [
|
|
{
|
|
type: 'text',
|
|
text: bodies.join('\n\n---\n\n'),
|
|
cache_control: { type: 'ephemeral', ttl },
|
|
},
|
|
],
|
|
messages: [{ role: 'user', content: 'warmup' }],
|
|
})
|
|
} catch {
|
|
// Fire-and-forget: pre-warm failure must never block the composer.
|
|
}
|
|
}
|