Harden machine handler capture

[SandroMaglione]

Summary

• Capture event dispatch maps, transition descriptors, and declared target arrays when handlers are registered so caller-owned mutation cannot make planner strategies diverge.
• Make indexed execution eligibility fail closed: unknown state semantics now select the generic planner until explicitly supported.
• Add public and internal regression coverage, plus concise Effect-internals testing, performance, and quality guidance for future agents.

Changeset

• Added or updated for a library or package-metadata change
• Not required because this PR does not change src/ or package.json

Validation

• pnpm check
• Relevant example checks, when examples changed (not applicable; no examples changed)
• Reviewed the automated type-performance report, when the public TypeScript API or inference changed
• Reviewed the automated runtime-performance report, when runtime behavior changed
• pnpm perf:types
• pnpm perf:runtime

Type instantiations are unchanged in every measured scenario. Runtime throughput and memory remain within the benchmark suite's expected run-to-run variability.

[github-actions]

Type performance

Measured with TypeScript 6.0.3 and skipLibCheck=true.

Scenario	Base	PR	Difference
Effect only	55	55	0 (0.0%)
Import effect-machine	55	55	0 (0.0%)
Machine.defineStates (3 states)	2,965	2,965	0 (0.0%)
Machine.make (3 states, 2 events)	8,715	8,715	0 (0.0%)
machine.handle (3 states, 2 transitions)	23,492	23,492	0 (0.0%)
machine.handle (depth 8)	117,075	117,075	0 (0.0%)
machine.handle (depth 12)	129,713	129,713	0 (0.0%)
machine.handle (depth 16)	145,055	145,055	0 (0.0%)
machine.handle (depth 24)	183,851	183,851	0 (0.0%)
machine.handle (wide depth 16)	214,189	214,189	0 (0.0%)
machine.handle (parallel/history/choice)	132,667	132,667	0 (0.0%)
machine.handle (4 successive calls)	128,674	128,674	0 (0.0%)
machine exact input/output/error/services	110,584	110,584	0 (0.0%)
execution adapter readiness	127,854	127,854	0 (0.0%)

Marginal instantiations are measured against the matching setup without that API call:

Scenario	Base	PR	Difference
Import effect-machine	0	0	0
Machine.defineStates (3 states)	2,910	2,910	0 (0.0%)
Machine.make (3 states, 2 events)	5,742	5,742	0 (0.0%)
machine.handle (3 states, 2 transitions)	14,777	14,777	0 (0.0%)
machine.handle (depth 8)	109,601	109,601	0 (0.0%)
machine.handle (depth 12)	121,039	121,039	0 (0.0%)
machine.handle (depth 16)	135,181	135,181	0 (0.0%)
machine.handle (depth 24)	171,577	171,577	0 (0.0%)
machine.handle (wide depth 16)	202,224	202,224	0 (0.0%)
machine.handle (parallel/history/choice)	114,331	114,331	0 (0.0%)
machine.handle (4 successive calls)	113,703	113,703	0 (0.0%)
machine exact input/output/error/services	100,670	100,670	0 (0.0%)
execution adapter readiness	102,092	102,092	0 (0.0%)

Check times (informational)

Scenario	Base	PR
Effect only	0.03s	0.03s
Import effect-machine	0.03s	0.03s
Machine.defineStates (3 states)	0.10s	0.11s
Machine.make (3 states, 2 events)	0.16s	0.17s
machine.handle (3 states, 2 transitions)	0.23s	0.25s
machine.handle (depth 8)	0.53s	0.53s
machine.handle (depth 12)	0.58s	0.58s
machine.handle (depth 16)	0.61s	0.60s
machine.handle (depth 24)	0.71s	0.68s
machine.handle (wide depth 16)	0.79s	0.77s
machine.handle (parallel/history/choice)	0.66s	0.64s
machine.handle (4 successive calls)	0.61s	0.62s
machine exact input/output/error/services	0.55s	0.52s
execution adapter readiness	0.62s	0.58s

Type instantiations are the comparison metric. Check time varies with runner load and is informational only.

[github-actions]

Runtime performance

Median of 5 independent benchmark processes on INTEL(R) XEON(R) PLATINUM 8573C with Node v24.18.0.

Pull request baseline

Scenario	Effect Machine	XState 5	XState 6 alpha
Plan counter transitions	151,864 transitions/s	149,257 transitions/s	17,535 transitions/s
Drain burst with terminal fence	659,681 increments/s	421,360 increments/s	190,841 increments/s
Drain burst with a change observer	608,604 increments/s	417,902 increments/s	191,315 increments/s
Lookup and send to one child	531,677 increments/s	422,148 increments/s	184,848 increments/s
Start and stop a machine	207,814 machines/s	209,380 machines/s	164,069 machines/s
Start and stop a parent with one child	56,786 families/s	80,587 families/s	53,605 families/s
Plan transitions through a compound state	136,961 transitions/s	—	—
Plan transitions through parallel regions	107,919 transitions/s	—	—
Drain burst through a compound state	609,105 events/s	370,190 events/s	192,293 events/s
Drain burst through two parallel regions	560,575 events/s	292,401 events/s	153,541 events/s
Drain a compound-state burst with a change observer	558,591 events/s	—	—

Effect runtime reference points

Scenario	Effect Machine	Effect runtime primitives
Start and stop a raw generic process	33,792 processes/s	—
Start and stop a raw compiled process	88,292 processes/s	—
Start and interrupt a suspended fiber	—	193,874 fibers/s
Start and stop a queue worker	—	141,764 workers/s
Start and stop an actor shell	—	130,804 actors/s
Start and stop two actor shells	—	81,473 families/s
Update an owner-only mutable snapshot	—	302,023,558 updates/s
Update a synchronized snapshot	—	1,764,885 updates/s
Create, resolve, and await a terminal latch	—	1,713,655 latches/s

Memory profile	Effect Machine	XState 5	XState 6 alpha	Effect runtime primitives
Idle machine	1.5 KiB	3.8 KiB	2.2 KiB	—
Raw generic managed process	12.1 KiB	—	—	—
Raw compiled process	2.6 KiB	—	—	—
Two independent idle machines	2.9 KiB	7.3 KiB	3.9 KiB	—
Idle parent with one child	4.8 KiB	5.5 KiB	4.0 KiB	—
Parent with observed child registry	9.0 KiB	—	—	—
Parent with observed invoked child snapshots	5.0 KiB	—	—	—
Suspended Effect fiber	—	—	—	0.7 KiB
Effect queue with waiting fiber	—	—	—	2.8 KiB
Effect mailbox actor shell	—	—	—	3.2 KiB
Two Effect actor shells	—	—	—	6.8 KiB

Effect Machine change from base

Metric	Base	Base variability	PR	PR variability	Difference
Plan counter transitions	147,249 transitions/s	2.5% MAD	151,864 transitions/s	1.9% MAD	+3.1%
Drain burst with terminal fence	618,266 increments/s	4.7% MAD	659,681 increments/s	0.9% MAD	+6.7%
Drain burst with a change observer	570,577 increments/s	1.8% MAD	608,604 increments/s	0.6% MAD	+6.7%
Lookup and send to one child	524,167 increments/s	1.0% MAD	531,677 increments/s	0.3% MAD	+1.4%
Start and stop a machine	210,881 machines/s	0.8% MAD	207,814 machines/s	1.7% MAD	-1.5%
Start and stop a parent with one child	58,817 families/s	2.7% MAD	56,786 families/s	1.0% MAD	-3.5%
Plan transitions through a compound state	138,462 transitions/s	2.1% MAD	136,961 transitions/s	3.0% MAD	-1.1%
Plan transitions through parallel regions	106,340 transitions/s	0.2% MAD	107,919 transitions/s	0.8% MAD	+1.5%
Drain burst through a compound state	614,142 events/s	0.6% MAD	609,105 events/s	0.9% MAD	-0.8%
Drain burst through two parallel regions	570,206 events/s	1.5% MAD	560,575 events/s	0.3% MAD	-1.7%
Drain a compound-state burst with a change observer	575,259 events/s	1.8% MAD	558,591 events/s	1.3% MAD	-2.9%
Idle machine heap per unit	1.5 KiB	0.1% MAD	1.5 KiB	0.0% MAD	-0.2%
Raw generic managed process heap per unit	12.1 KiB	0.0% MAD	12.1 KiB	0.0% MAD	+0.0%
Raw compiled process heap per unit	2.6 KiB	0.1% MAD	2.6 KiB	0.4% MAD	+0.6%
Two independent idle machines heap per unit	2.9 KiB	0.0% MAD	2.9 KiB	0.2% MAD	+0.2%
Idle parent with one child heap per unit	4.8 KiB	0.0% MAD	4.8 KiB	0.0% MAD	-0.1%
Parent with observed child registry heap per unit	9.0 KiB	0.0% MAD	9.0 KiB	0.0% MAD	+0.0%
Parent with observed invoked child snapshots heap per unit	5.0 KiB	0.2% MAD	5.0 KiB	0.0% MAD	+0.1%

Effect runtime reference change from base

Metric	Base	Base variability	PR	PR variability	Difference
Start and stop a raw generic process	33,105 processes/s	1.1% MAD	33,792 processes/s	0.6% MAD	+2.1%
Start and stop a raw compiled process	87,773 processes/s	0.3% MAD	88,292 processes/s	0.7% MAD	+0.6%
Start and interrupt a suspended fiber	191,718 fibers/s	0.6% MAD	193,874 fibers/s	0.5% MAD	+1.1%
Start and stop a queue worker	140,371 workers/s	1.7% MAD	141,764 workers/s	0.8% MAD	+1.0%
Start and stop an actor shell	129,870 actors/s	1.0% MAD	130,804 actors/s	0.6% MAD	+0.7%
Start and stop two actor shells	75,614 families/s	2.2% MAD	81,473 families/s	1.2% MAD	+7.7%
Update an owner-only mutable snapshot	302,114,804 updates/s	0.1% MAD	302,023,558 updates/s	0.0% MAD	-0.0%
Update a synchronized snapshot	1,801,211 updates/s	1.3% MAD	1,764,885 updates/s	3.3% MAD	-2.0%
Create, resolve, and await a terminal latch	1,729,571 latches/s	1.0% MAD	1,713,655 latches/s	1.4% MAD	-0.9%

Versions and interpretation

• Effect Machine: 0.3.0
• XState 5: 5.32.5
• XState 6 alpha: 6.0.0-alpha.31
• Effect runtime primitives: 4.0.0-beta.107

Higher throughput is better; lower heap is better. Variability is the median absolute deviation across independent processes, relative to their median. Runtime measurements on shared GitHub-hosted hardware remain informational, so small differences should be confirmed across multiple workflow runs.