Merge master into tool-result pruning

Retarget the pruning feature onto the merged compaction foundation.

Adapt content-only tool-result rewrite validation to the session-owned surface manager, migrate the demo wiring to repl-agent, and refresh generated type, catalog, and website contracts while preserving pressure and overflow pruning behavior.
This commit is contained in:
Tianyi Cui
2026-07-19 17:54:55 +08:00
972 changed files with 59334 additions and 11465 deletions

View File

@@ -48,7 +48,7 @@ describe('session-log invariants', () => {
session.append('user/message', { content: [{ type: 'text', text: 'hi' }], source: { kind: 'user' } }, { surfaceOp: 'append' })
session.append('step/start', { turn: 1, step: 1 })
session.append('assistant/chunk', { turn: 1, step: 1, chunk: { type: 'text-delta', index: 0, text: 'h' } })
session.append('assistant/message', { turn: 1, step: 1, content: [{ type: 'tool-call', id: CallId('c1'), name: 'echo', arguments: '{}' }] }, { surfaceOp: 'append' })
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [{ type: 'tool-call', id: CallId('c1'), name: 'echo', arguments: '{}' }] }, { surfaceOp: 'append' })
session.append('tool/call', { turn: 1, step: 1, callId: CallId('c1'), name: 'echo', arguments: '{}' })
session.append('tool/result', { turn: 1, step: 1, callId: CallId('c1'), content: [{ type: 'text', text: 'ok' }], isError: false }, { surfaceOp: 'append' })
session.append('step/end', { turn: 1, step: 1 })
@@ -208,7 +208,7 @@ describe('session-log invariants', () => {
expect(() => {
session.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
session.append('step/start', { turn: 1, step: 1 })
session.append('assistant/message', { turn: 1, step: 1, content: [
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [
{ type: 'tool-call', id: CallId('crashed'), name: 'bash', arguments: '{}' },
] }, { surfaceOp: 'append' })
session.append('tool/result', {
@@ -264,10 +264,10 @@ describe('session-log invariants', () => {
expect(() => {
session.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
session.append('step/start', { turn: 1, step: 1 })
session.append('assistant/message', { turn: 1, step: 1, content: [] }, { surfaceOp: 'append' })
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [] }, { surfaceOp: 'append' })
session.append('step/end', { turn: 1, step: 1 })
session.append('step/start', { turn: 1, step: 2 })
session.append('assistant/message', { turn: 1, step: 2, content: [] }, { surfaceOp: 'append' })
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 2, content: [] }, { surfaceOp: 'append' })
session.append('step/end', { turn: 1, step: 2 })
session.append('turn/end', { turn: 1, reason: { kind: 'completed' } })
session.append('turn/start', { turn: 2, trigger: { kind: 'message', source: { kind: 'user' } } })
@@ -329,7 +329,7 @@ describe('session-log invariants', () => {
const session = ctx.sessions.create()
session.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
session.append('step/start', { turn: 1, step: 1 })
expect(() => session.append('assistant/message', { turn: 1, step: 2, content: [] }, { surfaceOp: 'append' }))
expect(() => session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 2, content: [] }, { surfaceOp: 'append' }))
.toThrow(/open is turn 1\/step 1/)
})
})
@@ -477,8 +477,8 @@ describe('HMR safety', () => {
})
})
describe('surface invariants', () => {
async function toolResultRewriteFixture() {
describe('surface contract under the invariants composition', () => {
async function toolResultRewriteFixture(openRewriteTurn = true) {
const { ctx } = await setup()
const session = ctx.sessions.create()
session.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
@@ -507,7 +507,9 @@ describe('surface invariants', () => {
const original = session.append('tool/result', originalData, { surfaceOp: 'append' })
session.append('step/end', { turn: 1, step: 1 })
session.append('turn/end', { turn: 1, reason: { kind: 'completed' } })
session.append('turn/start', { turn: 2, trigger: { kind: 'message', source: { kind: 'user' } } })
if (openRewriteTurn) {
session.append('turn/start', { turn: 2, trigger: { kind: 'message', source: { kind: 'user' } } })
}
return { session, unrelated, original }
}
@@ -519,7 +521,7 @@ describe('surface invariants', () => {
session.append('step/start', { turn: 1, step: 1 })
expect(() => {
session.append('user/message', { content: [{ type: 'text', text: 'hi' }], source: { kind: 'user' } }, { surfaceOp: 'append' })
session.append('assistant/message', { turn: 1, step: 1, content: [] }, { surfaceOp: 'append', sourceEventSeqs: [1] })
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [] }, { surfaceOp: 'append', sourceEventSeqs: [1] })
}).not.toThrow()
})
@@ -529,7 +531,7 @@ describe('surface invariants', () => {
session.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
session.append('step/start', { turn: 1, step: 1 })
session.append('user/message', { content: [{ type: 'text', text: 'a' }], source: { kind: 'user' } }, { surfaceOp: 'append' })
session.append('assistant/message', { turn: 1, step: 1, content: [] }, { surfaceOp: { op: 'replace', start: 2, end: 2 }, sourceEventSeqs: [2] })
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [] }, { surfaceOp: { op: 'replace', start: 2, end: 2 }, sourceEventSeqs: [2] })
// no throw — well-formed replace op
})
@@ -545,6 +547,18 @@ describe('surface invariants', () => {
})).not.toThrow()
})
it('rejects a tool-result replacement outside a turn', async () => {
const { session, original } = await toolResultRewriteFixture(false)
expect(() => session.append('tool/result', {
...original.data,
content: [{ type: 'text', text: 'pruned' }],
}, {
surfaceOp: { op: 'replace', start: original.seq, end: original.seq },
sourceEventSeqs: [original.seq],
})).toThrow(/outside any open turn/)
})
it('rejects a tool-result replacement targeting an unrelated current node', async () => {
const { session, unrelated, original } = await toolResultRewriteFixture()
expect(() => session.append('tool/result', {
@@ -592,7 +606,7 @@ describe('surface invariants', () => {
session.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
session.append('step/start', { turn: 1, step: 1 })
expect(() => {
session.append('assistant/message', { turn: 1, step: 1, content: [] }, { surfaceOp: 'append', sourceEventSeqs: [] })
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [] }, { surfaceOp: 'append', sourceEventSeqs: [] })
}).not.toThrow()
expect(() => {
session.append('user/message', { content: [], source: { kind: 'user' } }, { surfaceOp: 'append', sourceEventSeqs: [] })
@@ -605,7 +619,7 @@ describe('surface invariants', () => {
session.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
session.append('user/message', { content: [{ type: 'text', text: 'a' }], source: { kind: 'user' } }, { surfaceOp: 'append' })
expect(() => {
session.append('assistant/message', { turn: 1, step: 1, content: [] }, { surfaceOp: 'append', sourceEventSeqs: [1, 1] })
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [] }, { surfaceOp: 'append', sourceEventSeqs: [1, 1] })
}).toThrow(/must not contain duplicates/)
})
@@ -616,19 +630,20 @@ describe('surface invariants', () => {
// The next event is seq 1. Referencing its own seq fails on "must reference
// earlier events" (the check order is: earlier first, then unknown).
expect(() => {
session.append('assistant/message', { turn: 1, step: 1, content: [] }, { surfaceOp: 'append', sourceEventSeqs: [1] })
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [] }, { surfaceOp: 'append', sourceEventSeqs: [1] })
}).toThrow(/must reference earlier/)
})
it('accepts sourceEventSeqs referencing a valid earlier event', async () => {
// Positive test: ref < current seq and ref is in knownSeqs → passes.
// Session seqs are contiguous, so every non-negative ref below the current
// seq necessarily names an existing earlier event.
const { ctx } = await setup()
const session = ctx.sessions.create()
session.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
session.append('step/start', { turn: 1, step: 1 })
// seqs so far: 0, 1. The next event at seq 2 references seq 1 → valid.
expect(() => {
session.append('assistant/message', { turn: 1, step: 1, content: [] }, { surfaceOp: 'append', sourceEventSeqs: [1] })
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [] }, { surfaceOp: 'append', sourceEventSeqs: [1] })
}).not.toThrow()
})
@@ -637,27 +652,10 @@ describe('surface invariants', () => {
const session = ctx.sessions.create()
session.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
expect(() => {
session.append('assistant/message', { turn: 1, step: 1, content: [] }, { surfaceOp: 'append', sourceEventSeqs: [99] })
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [] }, { surfaceOp: 'append', sourceEventSeqs: [99] })
}).toThrow(/must reference earlier/)
})
it('rejects sourceEventSeqs referencing unknown seq (gap in event log)', async () => {
// Create an impossible-through-public-API gap so seq 2 is earlier but unknown.
const { ctx } = await setup()
const session = ctx.sessions.create()
session.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
session.append('step/start', { turn: 1, step: 1 })
;(session as unknown as { log: unknown[] }).log.push({
type: 'assistant/chunk',
seq: 3,
time: Date.now(),
data: { turn: 1, step: 1, chunk: { type: 'text-delta', index: 0, text: 'x' } },
})
expect(() => {
session.append('assistant/message', { turn: 1, step: 1, content: [] }, { surfaceOp: 'append', sourceEventSeqs: [2] })
}).toThrow(/unknown seq 2/)
})
it('rejects a replace whose start is positioned after its end on the surface', async () => {
const { ctx } = await setup()
const session = ctx.sessions.create()
@@ -667,8 +665,8 @@ describe('surface invariants', () => {
session.append('user/message', { content: [{ type: 'text', text: 'b' }], source: { kind: 'user' } }, { surfaceOp: 'append' }) // seq 3
// Reversed range: start seq 3 is at a later surface position than end seq 2.
expect(() => {
session.append('assistant/message', { turn: 1, step: 1, content: [] }, { surfaceOp: { op: 'replace', start: 3, end: 2 }, sourceEventSeqs: [2, 3] })
}).toThrow(/is after end seq 2 .* on the surface/)
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [] }, { surfaceOp: { op: 'replace', start: 3, end: 2 }, sourceEventSeqs: [2, 3] })
}).toThrow(/is after end seq 2/)
})
it('rejects a replace whose sourceEventSeqs omits a shadowed surface node', async () => {
@@ -680,7 +678,7 @@ describe('surface invariants', () => {
session.append('user/message', { content: [{ type: 'text', text: 'b' }], source: { kind: 'user' } }, { surfaceOp: 'append' }) // seq 3
// Replace shadows surface nodes [2, 3] but records provenance for only [2].
expect(() => {
session.append('assistant/message', { turn: 1, step: 1, content: [{ type: 'text', text: 'sum' }] }, { surfaceOp: { op: 'replace', start: 2, end: 3 }, sourceEventSeqs: [2] })
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [{ type: 'text', text: 'sum' }] }, { surfaceOp: { op: 'replace', start: 2, end: 3 }, sourceEventSeqs: [2] })
}).toThrow(/must include every shadowed surface node; missing 3/)
})
@@ -692,7 +690,7 @@ describe('surface invariants', () => {
session.append('user/message', { content: [{ type: 'text', text: 'a' }], source: { kind: 'user' } }, { surfaceOp: 'append' }) // seq 2
session.append('user/message', { content: [{ type: 'text', text: 'b' }], source: { kind: 'user' } }, { surfaceOp: 'append' }) // seq 3
expect(() => {
session.append('assistant/message', { turn: 1, step: 1, content: [{ type: 'text', text: 'sum' }] }, { surfaceOp: { op: 'replace', start: 2, end: 3 }, sourceEventSeqs: [2, 3] })
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [{ type: 'text', text: 'sum' }] }, { surfaceOp: { op: 'replace', start: 2, end: 3 }, sourceEventSeqs: [2, 3] })
}).not.toThrow()
})
@@ -704,8 +702,8 @@ describe('surface invariants', () => {
session.append('user/message', { content: [{ type: 'text', text: 'a' }], source: { kind: 'user' } }, { surfaceOp: 'append' }) // seq 2
// seq 1 (step/start) is a real earlier event but never entered the surface.
expect(() => {
session.append('assistant/message', { turn: 1, step: 1, content: [] }, { surfaceOp: { op: 'replace', start: 1, end: 2 }, sourceEventSeqs: [1, 2] })
}).toThrow(/start seq 1 is not on the surface/)
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [] }, { surfaceOp: { op: 'replace', start: 1, end: 2 }, sourceEventSeqs: [1, 2] })
}).toThrow(/start seq 1 not found in surface/)
})
it('rejects a replace naming an end seq that is not on the surface', async () => {
@@ -716,8 +714,8 @@ describe('surface invariants', () => {
session.append('user/message', { content: [{ type: 'text', text: 'a' }], source: { kind: 'user' } }, { surfaceOp: 'append' }) // seq 2
// start (2) is on the surface but end (99) never entered it.
expect(() => {
session.append('assistant/message', { turn: 1, step: 1, content: [] }, { surfaceOp: { op: 'replace', start: 2, end: 99 }, sourceEventSeqs: [2] })
}).toThrow(/end seq 99 is not on the surface/)
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [] }, { surfaceOp: { op: 'replace', start: 2, end: 99 }, sourceEventSeqs: [2] })
}).toThrow(/end seq 99 not found in surface/)
})
it('rejects a replace whose range is reversed in surface position after a prior replace reordered it', async () => {
@@ -728,13 +726,13 @@ describe('surface invariants', () => {
session.append('user/message', { content: [{ type: 'text', text: 'a' }], source: { kind: 'user' } }, { surfaceOp: 'append' }) // seq 2
session.append('user/message', { content: [{ type: 'text', text: 'b' }], source: { kind: 'user' } }, { surfaceOp: 'append' }) // seq 3
// Replace node 2 (position 0) with seq 4 — surface is now [4, 3], so seq 4
// precedes seq 3 in linked-list order even though 4 > 3 numerically.
session.append('assistant/message', { turn: 1, step: 1, content: [{ type: 'text', text: 's' }] }, { surfaceOp: { op: 'replace', start: 2, end: 2 }, sourceEventSeqs: [2] }) // seq 4
// precedes seq 3 in surface order even though 4 > 3 numerically.
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [{ type: 'text', text: 's' }] }, { surfaceOp: { op: 'replace', start: 2, end: 2 }, sourceEventSeqs: [2] }) // seq 4
// A replace with start=3, end=4 passes the seq check (3 <= 4) but is
// reversed positionally (3 is at pos 1, 4 is at pos 0).
expect(() => {
session.append('assistant/message', { turn: 1, step: 1, content: [] }, { surfaceOp: { op: 'replace', start: 3, end: 4 }, sourceEventSeqs: [3, 4] }) // seq 5
}).toThrow(/is after end seq 4 .* on the surface/)
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [] }, { surfaceOp: { op: 'replace', start: 3, end: 4 }, sourceEventSeqs: [3, 4] }) // seq 5
}).toThrow(/is after end seq 4/)
})
it('accepts a replace whose start seq exceeds its end seq when the surface position order is valid', async () => {
@@ -748,9 +746,9 @@ describe('surface invariants', () => {
// head seq (4) is numerically GREATER than the tail seq (3): the surface is
// not seq-ordered. A replace spanning start=4 (pos 0) … end=3 (pos 1) is
// valid positionally and must be accepted even though start seq > end seq.
session.append('assistant/message', { turn: 1, step: 1, content: [{ type: 'text', text: 's' }] }, { surfaceOp: { op: 'replace', start: 2, end: 2 }, sourceEventSeqs: [2] }) // seq 4
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [{ type: 'text', text: 's' }] }, { surfaceOp: { op: 'replace', start: 2, end: 2 }, sourceEventSeqs: [2] }) // seq 4
expect(() => {
session.append('assistant/message', { turn: 1, step: 1, content: [] }, { surfaceOp: { op: 'replace', start: 4, end: 3 }, sourceEventSeqs: [4, 3] }) // seq 5
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [] }, { surfaceOp: { op: 'replace', start: 4, end: 3 }, sourceEventSeqs: [4, 3] }) // seq 5
}).not.toThrow()
})
@@ -762,7 +760,7 @@ describe('surface invariants', () => {
session.append('user/message', { content: [{ type: 'text', text: 'a' }], source: { kind: 'user' } }, { surfaceOp: 'append' }) // seq 2
// A replace with no sourceEventSeqs records no provenance for the node it shadows.
expect(() => {
session.append('assistant/message', { turn: 1, step: 1, content: [] }, { surfaceOp: { op: 'replace', start: 2, end: 2 } })
session.append('assistant/message', { provenance: { provider: 'mock', model: 'mock' }, turn: 1, step: 1, content: [] }, { surfaceOp: { op: 'replace', start: 2, end: 2 } })
}).toThrow(/must include every shadowed surface node; missing 2/)
})
@@ -773,30 +771,11 @@ describe('surface invariants', () => {
{ type: 'step/start' as const, seq: 1, time: 0, data: { turn: 1, step: 1 } },
{ type: 'user/message' as const, seq: 2, time: 0, data: { content: [{ type: 'text' as const, text: 'a' }], source: { kind: 'user' as const } }, surfaceOp: 'append' as const },
{ type: 'user/message' as const, seq: 3, time: 0, data: { content: [{ type: 'text' as const, text: 'b' }], source: { kind: 'user' as const } }, surfaceOp: 'append' as const },
{ type: 'assistant/message' as const, seq: 4, time: 0, data: { turn: 1, step: 1, content: [{ type: 'text' as const, text: 'sum' }] }, surfaceOp: { op: 'replace' as const, start: 2, end: 3 }, sourceEventSeqs: [2] },
{ type: 'assistant/message' as const, seq: 4, time: 0, data: { turn: 1, step: 1, content: [{ type: 'text' as const, text: 'sum' }], provenance: { provider: 'mock', model: 'mock' } }, surfaceOp: { op: 'replace' as const, start: 2, end: 3 }, sourceEventSeqs: [2] },
]
expect(() => ctx.sessions.create(undefined, { seed: badSeed })).toThrow(/must include every shadowed surface node; missing 3/)
})
it('rejects sourceEventSeqs on a non-surface event', async () => {
const { ctx } = await setup()
const session = ctx.sessions.create()
session.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
// Session rejects this at its own acceptance boundary. Emit a hand-built
// record to cover the listener's defensive check for alternate producers.
const event = { type: 'turn/end', seq: 1, time: 1, data: { turn: 1, reason: { kind: 'completed' } }, sourceEventSeqs: [0] }
expect(() => { ctx.emit(scopeTarget(session, undefined), 'session/event', session, event as never) })
.toThrow(/cannot carry sourceEventSeqs/)
})
it('rejects surfaceOp on a non-surface event', async () => {
const { ctx } = await setup()
const session = ctx.sessions.create()
session.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
const event = { type: 'turn/end', seq: 1, time: 1, data: { turn: 1, reason: { kind: 'completed' } }, surfaceOp: 'append' }
expect(() => { ctx.emit(scopeTarget(session, undefined), 'session/event', session, event as never) })
.toThrow(/cannot carry surfaceOp/)
})
})
describe('request-reconstruction cross-check (llm/stream)', () => {
@@ -808,7 +787,7 @@ describe('request-reconstruction cross-check (llm/stream)', () => {
session.append('user/message', { content: [{ type: 'text', text: 'hi' }], source: { kind: 'user' } }, { surfaceOp: 'append' })
const boundary = session.deriveMessages()
session.append('step/start', { turn: 1, step: 1 })
session.append('request/header', { header: { config: { model: 'm' } }, reason: 'initial' })
session.append('request/header', { header: { config: { provider: 'mock', model: 'm' } }, reason: 'initial' })
return { ctx, session, boundary }
}
@@ -838,7 +817,7 @@ describe('request-reconstruction cross-check (llm/stream)', () => {
it('expects the folded header\'s session prefix ahead of the derivation (prefix + derived)', async () => {
const { ctx, session, boundary } = await requestSetup()
const prefix = { role: 'user' as const, content: [{ type: 'text' as const, text: '<system-reminder>catalog</system-reminder>' }] }
session.append('request/header-delta', { messagePrefix: [prefix] })
session.append('request/header', { header: { config: { provider: 'mock', model: 'm' }, messagePrefix: [prefix] }, reason: 'change' })
// The prefixed request matches the fold…
const prefixed = Object.freeze({ model: 'm', messages: Object.freeze([prefix, ...boundary]), sessionId: session.id })
expect(() => { dispatch(ctx, prefixed) }).not.toThrow()
@@ -907,7 +886,7 @@ describe('request cross-check ordering (prepend)', () => {
session.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
session.append('user/message', { content: [{ type: 'text', text: 'hi' }], source: { kind: 'user' } }, { surfaceOp: 'append' })
session.append('step/start', { turn: 1, step: 1 })
session.append('request/header', { header: { config: { model: 'm' } }, reason: 'initial' })
session.append('request/header', { header: { config: { provider: 'mock', model: 'm' } }, reason: 'initial' })
const divergent = Object.freeze({
model: 'm',