Add live machine inspection

[SandroMaglione]

Summary

• Add Machine.prepare(machine).inspection, a hot root-scoped Effect Stream for ordered lifecycle, delivery, processing, emission, activity, and termination records across the local ownership tree.
• Preserve causal evidence and equivalent behavior in the generic and compact runtimes, including explicit child-to-parent sends, retained microsteps, Effect/timer cancellation, startup failures, and typed decode failures.
• Add AtomMachine.inspection, public documentation, runtime/type/differential coverage, and a pre-1.0 minor changeset.

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 (no example package changed)
• Automated type-performance measurement passed locally; CI comparison pending
• Automated runtime- and memory-performance measurement passed locally; CI comparison pending

[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)	3,039	3,039	0 (0.0%)
Machine.make (3 states, 2 events)	13,199	13,199	0 (0.0%)
machine.handle (3 states, 2 transitions)	30,443	30,443	0 (0.0%)
Machine.invoke (state-dependent Effect)	70,236	70,236	0 (0.0%)
machine.handle (depth 24)	188,212	188,212	0 (0.0%)
machine.handle (wide depth 16)	220,710	220,710	0 (0.0%)
machine.handle (parallel/history/choice)	128,069	128,069	0 (0.0%)
machine.handle (4 successive calls)	135,838	135,838	0 (0.0%)
machine exact input/output/error/services	109,952	109,952	0 (0.0%)
execution adapter readiness	121,683	121,683	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)	2,984	2,984	0 (0.0%)
Machine.make (3 states, 2 events)	10,152	10,152	0 (0.0%)
machine.handle (3 states, 2 transitions)	17,244	17,244	0 (0.0%)
Machine.invoke (state-dependent Effect)	59,203	59,203	0 (0.0%)
machine.handle (depth 24)	168,143	168,143	0 (0.0%)
machine.handle (wide depth 16)	201,702	201,702	0 (0.0%)
machine.handle (parallel/history/choice)	102,415	102,415	0 (0.0%)
machine.handle (4 successive calls)	114,587	114,587	0 (0.0%)
machine exact input/output/error/services	93,421	93,421	0 (0.0%)
execution adapter readiness	87,555	87,555	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	0.10s
Machine.make (3 states, 2 events)	0.17s	0.17s
machine.handle (3 states, 2 transitions)	0.25s	0.25s
Machine.invoke (state-dependent Effect)	0.40s	0.41s
machine.handle (depth 24)	0.66s	0.67s
machine.handle (wide depth 16)	0.68s	0.70s
machine.handle (parallel/history/choice)	0.54s	0.54s
machine.handle (4 successive calls)	0.54s	0.54s
machine exact input/output/error/services	0.51s	0.48s
execution adapter readiness	0.53s	0.50s

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 AMD EPYC 7763 64-Core Processor with Node v24.19.0.

Pull request baseline

Scenario	Effect Machine
Plan counter transitions	123,276 transitions/s
Drain burst with terminal fence	379,932 increments/s
Drain burst with a change observer	354,097 increments/s
Lookup and send to one child	311,575 increments/s
Start and stop a machine	146,349 machines/s
Start and stop a parent with one child	31,105 families/s
Plan transitions through a compound state	108,748 transitions/s
Plan transitions through parallel regions	86,674 transitions/s
Drain burst through a compound state	351,791 events/s
Drain burst through two parallel regions	324,232 events/s
Drain a compound-state burst with a change observer	329,250 events/s

Process runtime reference points

Scenario	Effect Machine
Start and stop a raw generic process	14,813 processes/s
Start and stop a raw compiled process	59,984 processes/s

Memory profile	Effect Machine
Idle machine	1.8 KiB
Raw generic managed process	13.9 KiB
Raw compiled process	3.1 KiB
Two independent idle machines	3.4 KiB
Idle parent with one child	5.5 KiB
Parent with observed child registry	9.7 KiB
Parent with observed invoked child snapshots	6.0 KiB

Effect Machine change from base

Metric	Base	Base variability	PR	PR variability	Difference
Plan counter transitions	122,118 transitions/s	0.2% MAD	123,276 transitions/s	0.7% MAD	+0.9%
Drain burst with terminal fence	380,268 increments/s	2.1% MAD	379,932 increments/s	0.5% MAD	-0.1%
Drain burst with a change observer	354,489 increments/s	1.2% MAD	354,097 increments/s	0.8% MAD	-0.1%
Lookup and send to one child	313,011 increments/s	2.1% MAD	311,575 increments/s	0.6% MAD	-0.5%
Start and stop a machine	153,327 machines/s	0.3% MAD	146,349 machines/s	1.5% MAD	-4.6%
Start and stop a parent with one child	33,117 families/s	4.3% MAD	31,105 families/s	3.4% MAD	-6.1%
Plan transitions through a compound state	109,835 transitions/s	2.2% MAD	108,748 transitions/s	0.8% MAD	-1.0%
Plan transitions through parallel regions	86,428 transitions/s	1.5% MAD	86,674 transitions/s	0.8% MAD	+0.3%
Drain burst through a compound state	371,599 events/s	1.5% MAD	351,791 events/s	0.9% MAD	-5.3%
Drain burst through two parallel regions	349,294 events/s	0.6% MAD	324,232 events/s	0.9% MAD	-7.2%
Drain a compound-state burst with a change observer	349,333 events/s	0.6% MAD	329,250 events/s	1.2% MAD	-5.7%
Idle machine heap per unit	1.7 KiB	0.0% MAD	1.8 KiB	0.1% MAD	+4.0%
Raw generic managed process heap per unit	13.7 KiB	0.0% MAD	13.9 KiB	0.0% MAD	+1.1%
Raw compiled process heap per unit	2.9 KiB	0.1% MAD	3.1 KiB	0.2% MAD	+6.0%
Two independent idle machines heap per unit	3.3 KiB	0.0% MAD	3.4 KiB	0.0% MAD	+4.1%
Idle parent with one child heap per unit	5.2 KiB	0.0% MAD	5.5 KiB	0.0% MAD	+5.2%
Parent with observed child registry heap per unit	9.4 KiB	0.0% MAD	9.7 KiB	0.0% MAD	+2.9%
Parent with observed invoked child snapshots heap per unit	5.6 KiB	0.1% MAD	6.0 KiB	0.0% MAD	+7.0%

Process runtime reference change from base

Metric	Base	Base variability	PR	PR variability	Difference
Start and stop a raw generic process	15,581 processes/s	0.8% MAD	14,813 processes/s	0.4% MAD	-4.9%
Start and stop a raw compiled process	62,775 processes/s	0.5% MAD	59,984 processes/s	1.0% MAD	-4.4%

Regression guard

No large, noise-adjusted throughput or heap regressions detected.

Versions and interpretation

• Effect Machine: 0.12.0

Higher throughput is better; lower heap is better. Variability is the median absolute deviation across independent processes, relative to their median. Small differences on shared GitHub-hosted hardware remain informational; the required guard rejects only large changes beyond the measured noise allowance.