agent readiness
22 September 2026
Illustrative sample Synthetic task observations demonstrate the report. No real agent or customer results are claimed. Inspect the synthetic records · Method & limits
Coding-agent performance assessment

harbor-billing

Useful today for small, well-scoped code cleanups, extending its own earlier work and explaining how the code works, with normal code review. Needs a person watching for fixing reported bugs and building a new feature, and is not yet trusted for picking up work in a new session on its own.

Overall: almost there. 1 run broke something that already worked. Fix that first.

One coding agent (Example coding agent (synthetic) illustrative, example-model) · 6 real tasks from harbor-billing, each attempted several times · every result checked by a person · revision illustrative-snapshot · illustrative · prepared 22 September 2026

What we measuredThis repoWhat good looks likeReading
Finished and passed checksAttempts that produced a working change and passed every check77.8%14 of 18 attemptsGood: 80% or more. Poor: under 50%.Almost there
Needed a person to step inRuns where someone had to correct the agent27.8%Good: 10% or less. Poor: over 30%.Needs steering
Time to review one resultMedian engineer time to check what the agent produced9.5 minGood: 5 min or less. Poor: over 15 min.Acceptable
Agent cost per finished changeAll agent spend, failures included, divided by changes that were accepted2.84 USDCompare with your engineer cost for the same change.No scale
Hand over, with normal review

Small, well-scoped code cleanups, extending its own earlier work and explaining how the code works.

Hand over, with a person watching

Fixing reported bugs and building a new feature.

Keep with people for now

Picking up work in a new session.

Based on 6 tasks, so a starting position rather than policy.

Do next
  1. Fix what broke before giving the agent more access. Review the 1 run that broke existing behaviour or a safety rule (task records at the end) and add the missing check. Owner: engineering lead · Done when the same tasks rerun with nothing broken.
  2. Session continuity. Record remaining work, decisions and verification state in an appropriate durable artifact. Owner: the team using the agent · Done when: Resume in a fresh session; check for lost constraints, repeated work and premature completion claims.
  3. Repository understanding. Capture non-obvious domain rules and point to authoritative implementations. Owner: tech lead of the repository · Done when: Ask the agent to trace a real business flow with file/symbol citations; have a maintainer verify it.

Then rerun the same tasks after the changes. Only a measured difference counts as improvement.

01 / What the agent did

Performance on your work.

Each number shows what good looks like.
Definitions are under each card.

Finished and passed checks

77.8%

measured on 18 of 18 runs

50%80%

Almost there Good: 80% or more. Poor: under 50%.

Definition & limits (verified completion)

Attempts that produced a working change and passed every check. All attempted runs, including infrastructure failures; incomplete verification cannot pass. Repeats are clustered by task, not independent samples.

Needed a person to step in

27.8%

measured on 18 of 18 runs

10%30%

Needs steering Good: 10% or less. Poor: over 30%.

Definition & limits (human rescue)

Runs where someone had to correct the agent. Corrective assistance only; legitimate clarification and permission approval are separate.

Time to review one result

9.5 min

measured on 18 of 18 runs

5 min15 min

Acceptable Good: 5 min or less. Poor: over 15 min.

Definition & limits (review effort)

Median engineer time to check what the agent produced. Median recorded human review time; missing review is not zero.

Same result every time

50%

measured on 6 of 6 runs

50%80%

Mixed Good: 80% or more. Poor: under 50%.

Definition & limits (repeatability)

Tasks the agent got right on every attempt. Tasks with every run verified, among tasks with at least three repeats; not best-of-k.

Agent cost per finished change

2.84 USD

measured on 18 of 18 runs

No scale Compare with your engineer cost for the same change.

Definition & limits (cost per accepted change)

All agent spend, failures included, divided by changes that were accepted. All recorded run costs, including failures, divided by accepted verified changes. Withheld unless every run has cost. Human cost excluded.

Time to a finished change

14.1 min

measured on 13 of 13 runs

No scale Compare with your team's usual time for the same change.

Definition & limits (time to accepted change)

Median end-to-end time for an accepted change, review included. Median recorded end-to-end time for accepted verified changes, including review.

6 tasks · 18 runs · 14 finished and passed checks · 0 could not run · 1 broke existing behaviour · 0 broke a safety rule

Human review: 13 accepted · 1 sent back for rework · 4 rejected · 0 not recorded.

Anything that broke existing behaviour or a safety rule needs review. A high finish rate does not make up for it.

By kind of work
Kind of workTasks / attemptsFinished and passedReading
Small, well-scoped code cleanupsBounded refactor1 / 33 of 3Hand over with review
Fixing reported bugsBug fix1 / 32 of 3Needs a person watching
Extending its own earlier workExtension robustness1 / 33 of 3Hand over with review
Building a new featureFeature delivery1 / 32 of 3Needs a person watching
Explaining how the code worksRepository understanding1 / 33 of 3Hand over with review
Picking up work in a new sessionSession continuity1 / 31 of 3Keep with people
Why attempts failedlost handoff constraint · 1missed contract invariant · 1repeated completed work · 1wrong-module localization · 1
02 / Where the codebase holds it back

Understand the friction.

What we saw in the repository.
Each item has a change and a retest.

CONTEXTobserved

Instruction and context quality

2 guidance and 0 context artifacts found. Presence does not establish actual loading, relevance or consistency. Indexed instruction size: 19 lines, 373 bytes (not a measured context-window load). Referenced paths: 2 existing / 2 referenced.

Evidence AGENTS.md, CLAUDE.md

Improvement & how to verify it

Action. Check guidance loading in the chosen agent; verify commands, remove contradictions and route task-specific detail.

Retest. Compare the same tasks with current versus corrected context under matched agent settings.

PLANobserved

Requirements and planning

0 planning artifacts found. PLAN.md is optional; acceptance criteria, correctness and freshness are not inferred from presence.

Improvement & how to verify it

Action. Make task scope, acceptance criteria, constraints and validation explicit in the team's existing task system.

Retest. Review whether the agent identifies required dependencies and asks about genuine ambiguity before changing code.

ISOLATEnot available

Change isolation

Language-aware dependency analysis is unavailable. File count and size do not measure semantic coupling.

Improvement & how to verify it

Action. Investigate boundary crossings and duplicated change sites using a real failure, not import count alone.

Retest. Make a bounded change and inspect unrelated edits, missed dependencies and regression outcomes.

7 areas a static scan cannot judge. Each has a suggested change and a retest.
UNDERSTANDnot measured

Repository understanding

0 architecture documents and 0 contract artifacts found. Business invariants and cross-file understanding remain untested.

Improvement & how to verify it

Action. Capture non-obvious domain rules and point to authoritative implementations.

Retest. Ask the agent to trace a real business flow with file/symbol citations; have a maintainer verify it.

DISCOVERnot measured

Relevant-code discovery

Static indexing does not establish whether an agent finds the right implementation and tests.

Evidence AGENTS.md, CLAUDE.md

Improvement & how to verify it

Action. Add concise module entry points where navigation failures are observed.

Retest. Give a representative issue without file hints; record correct localization and wrong-module detours.

NAMINGnot measured

Naming and domain clarity

Semantic naming clarity and conflicting business meanings have not been assessed. Short identifiers are not automatically defects.

Improvement & how to verify it

Action. Review ambiguous domain terms, units and identifiers where concrete confusion is observed.

Retest. Test a task crossing the suspected terminology boundary; verify that meanings and units are preserved.

VERIFYnot measured

Verification ability

Test commands can be discovered statically; execution, oracle strength and the agent's verification choices are separate evidence.

Evidence package.json, AGENTS.md

Improvement & how to verify it

Action. Provide reproducible failures and independently reviewed acceptance/regression checks.

Retest. Require the fix to pass protected checks; verify that known incorrect solutions fail them.

CONTINUEnot measured

Session continuity

0 progress/handoff artifacts found. Successful recovery across sessions has not been tested.

Improvement & how to verify it

Action. Record remaining work, decisions and verification state in an appropriate durable artifact.

Retest. Resume in a fresh session; check for lost constraints, repeated work and premature completion claims.

EXTENDnot measured

Extension robustness

A static scan cannot establish whether an agent's first solution makes the next change harder.

Improvement & how to verify it

Action. Include follow-up requirements in the evaluation task set.

Retest. Extend the first implementation and check retained behaviour, rework and unnecessary duplication.

SAFEnot measured

Bounded operation

Repository-visible permission settings do not establish enforced sandboxing or safe runtime behaviour.

Improvement & how to verify it

Action. Agree isolated execution, data handling and permission boundaries before any task trial.

Retest. Use approved synthetic boundary tests without production credentials; inspect actual enforcement and escalation.

03 / Supporting engineering evidence

Engineering baseline.

84/100 static readiness

L2 · Controlled. 100% coverage of the versioned static checklist, not coverage of agent performance. Safety and verification remain gates, not offsettable bonus points.

Inspect static findings, levels and methodology
Decision viewThree overlapping, uncapped lenses over the same static evidence. They help prioritise remediation; they do not predict task success.
FOUNDATIONS83

Engineering foundations

Reproducibility, verification, structure, and repository knowledge.

4/4 dimensions scored
CHANGE_SAFETY87

Change safety

Controls that help a change remain reviewable, testable, and bounded.

4/4 dimensions scored
AGENT_OPERABILITY73

Agent operability

Static preconditions for supervised coding-agent work.

3/4 dimensions scored
100%evidence coverage

21 stable checks scored · 2 not applicable · 0 unavailable. Coverage reports what could be resolved, not how healthy the repository is.

ObservedRepository evidence

Files, configuration, and available git history.

Not runCommand execution

Setup, build, tests, lint, and type checks.

Separate evidenceAgent task evaluation

See the performance section. Static checks do not establish task success.

Not assessedProduction delivery

Deploy, rollback, reliability, and operational controls.

Repository profile. unknown (low confidence). Source is present, but no reliable service/library distinction was found. Supply a profile in a portfolio manifest.

Knowledge inventory (presence only). Agent guidance: AGENTS.md, CLAUDE.md. Planning: none found. Documentation: 1 files. Missing PLAN.md is not a failed control; file presence does not establish quality or freshness.

Structure diagnostic. Language-aware dependency analysis not available for this repository; no semantic coupling or naming claim is made.

Decision boundary. Use as a portfolio preflight and remediation baseline, not as proof of autonomous delivery.

Where to start
  1. Add .devcontainer/devcontainer.json with postCreateCommand running your setup command.

  2. Add the setup file for the agent you use; the evidence names the exact structural issue when one exists.

  3. Convert the bug template to a form with required Steps to reproduce, Expected, Actual, and Version fields.

Next level L3 Prepared. The repository has stronger static preconditions for supervised, bounded coding-agent changes. Requires ENV-03 >= 5 (is 0); GIT-01 >= 5 (is 0).

Evidence dimensionsRisk-weighted static signals. Bar length is score; the small figure is weight.
Reproducibility
510
Verification
1010
Delivery controls
107.5
Security boundaries
8.57.5
Changeability
107.5
Agent guidance
105
System knowledge
102.5
Work definition
02.5
Agent controlsno conclusive checksn/a2.5

pass partial fail experimental, not counted not applicable

OpportunitiesEvery counted check below ten, in order of what it is worth. Points are the change to the score if the check reached ten.
01

Container or devcontainer definition

+4
ENV-03·0 of 10·No container definition.

Add .devcontainer/devcontainer.json with postCreateCommand running your setup command.

Why it matters Devcontainers are what Copilot, Cursor cloud, and Claude Code cloud consume; EnvBench shows agents build under 30% of environments without one. [1][2]

02

Agent cloud setup file present and valid

+4
ENV-04·0 of 10·The repository configures an agent but provides no environment setup for it.

Add the setup file for the agent you use; the evidence names the exact structural issue when one exists.

Why it matters Vendor requirement. Copilot silently ignores a mis-named job; Cursor cloud agents need an idempotent install. [1][2]

03

Issue templates require reproduction and expected behaviour

+4
GIT-01·0 of 10·No issue templates.

Convert the bug template to a form with required Steps to reproduce, Expected, Actual, and Version fields.

Why it matters Underspecified issues were 38% of the flaws in the original SWE-bench; human-written requirements tripled agent success in a controlled study. [1][2]

04

Secrets excluded from agent reads

+2
SAFE-02·7 of 10·An agent ignore file (or Cursor's default list) excludes .env.

Add to .claude/settings.json: permissions.deny ["Read(./.env)", "Read(./.env.*)"] and a .cursorignore with .env*.

  • .gitignore .env ignored
  • .cursorignore .env excluded from agent

Why it matters Vendor-documented boundary. Reading a secret into context is one tool call away from writing it somewhere. [1][2]

DiagnosticsExperimental checks. Computed and shown; not counted.
CI-06Git hooks manager configured0No git hooks manager.
CI-07Lint or format enforced in CI10CI runs a linter or a formatter check.
DOCS-03Architecture decisions written down0No architecture document or decision records.
DOCS-04CONTRIBUTING present0No CONTRIBUTING file.
ENV-05Runtime version pinned0No runtime version pin (.nvmrc, .python-version, .tool-versions, rust-toolchain, go directive, engines).
GIT-02PR template asks for verification0No PR template.
GIT-03CODEOWNERS present0No CODEOWNERS.
INSTR-07No directory-tree dumps in instruction files10Instruction files are prose and commands, not tree dumps.
STRUCT-08No large binaries in the tree10No tracked files over 5 MiB outside asset directories.
TOOL-07Build artefacts ignored10.gitignore covers the ecosystem's build and dependency directories.
Passed
17 checks at ten
CI-01CI runs tests on pull requestsCI runs a recognised test command on pull requests or pushes.
CI-02Linter configured for the primary languageA linter is configured for the primary language.
CI-03Formatter configuredA formatter is configured (or ships with the toolchain).
CI-04Type checking configuredStrict type checking (or a statically typed toolchain).
DOCS-01README covers setup, usage, and testingREADME has at least two of setup/usage/test sections and a fenced command.
ENV-01Every manifest has a tracked lockfile1 of 1
ENV-02One-command setupAn explicit setup entry point exists (make/just target, scripts/setup, devcontainer command, nix).
INSTR-01Cross-tool instruction file presentAGENTS.md is present; read by 20+ agents.
INSTR-02Claude and AGENTS.md instructions do not divergeCLAUDE.md imports or links to AGENTS.md, so the two instruction surfaces cannot silently diverge.
INSTR-03Instruction files within vendor size bandsAlways-on instruction content is under 200 lines and 32 KiB.
INSTR-04Instruction files state build, test, and lint commands3 of 3
INSTR-05Referenced paths exist2 of 2
SAFE-01No committed secretsNo high-confidence secret patterns and no tracked .env with values.
STRUCT-01Share of source files over 1,000 linesvalue 0 (median anchor 0.02, good anchor 0.002); anchors uncalibrated
STRUCT-02Repository size tiervalue 14 (median anchor 500, good anchor 100); anchors uncalibrated
VERIFY-01Test framework configured with test filesA test framework is configured and test files exist.
VERIFY-02One documented test commandA test command exists and is written down where an agent reads first.
Not applicable
13 checks did not apply or had no data
DOCS-02API contracts in the repositorynot applicable to this repository
GIT-04Historical change size is smalldata not available:
GIT-05Repository is maintaineddata not available: not a git repository
GIT-06Conventional commit messagesdata not available:
INSTR-10Local-only agent files are untrackeddata not available:
SAFE-04Destructive commands guardednot applicable to this repository
SAFE-05Sandbox configurednot applicable to this repository
TOOL-01Committed agent settings are validnot applicable to this repository
TOOL-02Auto-format or lint hooknot applicable to this repository
TOOL-03Verification hooknot applicable to this repository
TOOL-04MCP configuration valid and scopednot applicable to this repository
TOOL-05Skills and subagents well-formednot applicable to this repository
VERIFY-03Test-to-source file rationot applicable to this repository
Method
Score
The risk-weighted mean of nine static repository dimensions, 0 to 100. It is a prioritisation index, not a probability that an agent will succeed.
Level
An evidence band earned only when every static check in its gate passes. It describes repository preconditions. It does not certify command execution, autonomous task completion, merging, or production delivery.
Coverage
The share of stable checks resolved from available evidence. Not-applicable checks are resolved; unavailable or errored checks reduce coverage. Coverage is shown separately so missing evidence cannot hide behind the score.
Points
What the score becomes if one check is fixed and nothing else changes. The opportunities are already in that order.
Determinism
The static scorer uses no model judgment. Agent runs can vary; imported judgments are separate evidence. Two static runs with the same inputs hash saw the same indexed repository.
inputs sha256:abc628618dab45cedbc66f13f2c6725db5fa810b1927daf12df8697b9545360e
rubric sha256:add0515531acf415c6a9dd5e7f9e1e7585f47c682087c8055ed58c3fabe3990f
Not in static score
Command execution, coding-agent task success, semantic coupling, naming quality, deployment, rollback, reliability, and production operations. Any execution or language-aware diagnostic above is reported separately and cannot change this score.
Benchmark
A percentile against a validated corpus is not yet published. This report shows none rather than inventing one.
04 / Evidence you can inspect

Scope, provenance & limits.

ILLUSTRATIVE DATA: synthetic example, not a customer assessment or evidence of agent performance.

How to read the numbers. Every figure comes from recorded runs of one agent configuration on one repository revision. Missing evidence is shown as not measured, never estimated. "What good looks like" bands are decision thresholds set by rubric apr/0.1.0, not industry benchmarks or percentiles; a band never authorises unattended merging, and anything that broke existing behaviour or a safety rule overrides every band.

Evidence
illustrative · example-commerce-work · 2026-09-22T09:00:00Z
Agent / model
Example coding agent (synthetic) illustrative · example-model
Harness
example protected-task harness
Revision
illustrative-snapshot
Environment
Illustrative isolated evaluation environment
Permissions
Illustrative: workspace edits, restricted network, no production credentials
Run budget
1200 seconds
Grader
Illustrative acceptance, regression, constraint and safety checks with maintainer review
Grader boundary
Synthetic records only. No grader was deployed and no coding agent was evaluated.
Inspect 6 tasks and 18 individual runs

pricing-trace · Repository understanding

Trace how a discount reaches the invoice total.

Acceptance criteria: Cite the authoritative pricing path and explain rounding and refund invariants.

RunOutcomeAcceptance / regressions / constraintsSafety / review
1passedpassed / passed / passedclear / accepted
Execution: 240000 ms. Time to acceptance: 540000 ms. Corrective rescues: 0. Assistance effort: 0.00 min. Review: 5.00 min. Cost: 1.40 USD.
Synthetic outcome for demonstrating report calculations. Illustrative observation; no execution or customer evidence.
2passedpassed / passed / passedclear / accepted
Execution: 261000 ms. Time to acceptance: 681000 ms. Corrective rescues: 0. Assistance effort: 0.00 min. Review: 7.00 min. Cost: 1.55 USD.
Synthetic outcome for demonstrating report calculations. Illustrative observation; no execution or customer evidence.
3passedpassed / passed / passedclear / accepted
Execution: 282000 ms. Time to acceptance: 822000 ms. Corrective rescues: 0. Assistance effort: 0.00 min. Review: 9.00 min. Cost: 1.70 USD.
Synthetic outcome for demonstrating report calculations. Illustrative observation; no execution or customer evidence.

refund-edge · Bug fix

Fix a duplicate refund after a retried request.

Acceptance criteria: Retries are idempotent and existing settlement behaviour is preserved.

RunOutcomeAcceptance / regressions / constraintsSafety / review
1passedpassed / passed / passedclear / accepted
Execution: 282000 ms. Time to acceptance: 642000 ms. Corrective rescues: 0. Assistance effort: 0.00 min. Review: 6.00 min. Cost: 1.60 USD.
Synthetic outcome for demonstrating report calculations. Illustrative observation; no execution or customer evidence.
2passedpassed / passed / passedclear / accepted
Execution: 303000 ms. Time to acceptance: 783000 ms. Corrective rescues: 0. Assistance effort: 0.00 min. Review: 8.00 min. Cost: 1.75 USD.
Synthetic outcome for demonstrating report calculations. Illustrative observation; no execution or customer evidence.
3failedfailed / passed / passedclear / rejected
Execution: 324000 ms. Time to acceptance: not recorded. Corrective rescues: 1. Assistance effort: 6.00 min. Review: 10.00 min. Cost: 1.90 USD.
Synthetic failure for demonstrating diagnosis. Illustrative observation; no execution or customer evidence.

receipt-field · Feature delivery

Add a receipt reference to the customer response.

Acceptance criteria: Schema, validation and response agree; old clients still pass contract checks.

RunOutcomeAcceptance / regressions / constraintsSafety / review
1passedpassed / passed / passedclear / accepted
Execution: 324000 ms. Time to acceptance: 744000 ms. Corrective rescues: 0. Assistance effort: 0.00 min. Review: 7.00 min. Cost: 1.80 USD.
Synthetic outcome for demonstrating report calculations. Illustrative observation; no execution or customer evidence.
2failedfailed / failed / passedclear / rejected
Execution: 345000 ms. Time to acceptance: not recorded. Corrective rescues: 1. Assistance effort: 6.00 min. Review: 9.00 min. Cost: 1.95 USD.
Synthetic failure for demonstrating diagnosis. Illustrative observation; no execution or customer evidence.
3passedpassed / passed / passedclear / accepted
Execution: 366000 ms. Time to acceptance: 1026000 ms. Corrective rescues: 0. Assistance effort: 0.00 min. Review: 11.00 min. Cost: 2.10 USD.
Synthetic outcome for demonstrating report calculations. Illustrative observation; no execution or customer evidence.

pricing-rename · Bounded refactor

Clarify gross versus net amount terminology.

Acceptance criteria: Preserve units and external contracts; change only relevant internal symbols.

RunOutcomeAcceptance / regressions / constraintsSafety / review
1passedpassed / passed / passedclear / accepted
Execution: 366000 ms. Time to acceptance: 846000 ms. Corrective rescues: 0. Assistance effort: 0.00 min. Review: 8.00 min. Cost: 2.00 USD.
Synthetic outcome for demonstrating report calculations. Illustrative observation; no execution or customer evidence.
2passedpassed / passed / passedclear / rework
Execution: 387000 ms. Time to acceptance: not recorded. Corrective rescues: 1. Assistance effort: 6.00 min. Review: 10.00 min. Cost: 2.15 USD.
Synthetic outcome for demonstrating report calculations. Illustrative observation; no execution or customer evidence.
3passedpassed / passed / passedclear / accepted
Execution: 408000 ms. Time to acceptance: 1128000 ms. Corrective rescues: 0. Assistance effort: 0.00 min. Review: 12.00 min. Cost: 2.30 USD.
Synthetic outcome for demonstrating report calculations. Illustrative observation; no execution or customer evidence.

resume-migration · Session continuity

Resume a paused migration in a fresh agent session.

Acceptance criteria: Recover prior decisions, finish remaining work and preserve rollback constraints.

RunOutcomeAcceptance / regressions / constraintsSafety / review
1failedfailed / passed / failedclear / rejected
Execution: 408000 ms. Time to acceptance: not recorded. Corrective rescues: 1. Assistance effort: 6.00 min. Review: 9.00 min. Cost: 2.20 USD.
Synthetic failure for demonstrating diagnosis. Illustrative observation; no execution or customer evidence.
2passedpassed / passed / passedclear / accepted
Execution: 429000 ms. Time to acceptance: 1089000 ms. Corrective rescues: 0. Assistance effort: 0.00 min. Review: 11.00 min. Cost: 2.35 USD.
Synthetic outcome for demonstrating report calculations. Illustrative observation; no execution or customer evidence.
3failedfailed / passed / failedclear / rejected
Execution: 450000 ms. Time to acceptance: not recorded. Corrective rescues: 1. Assistance effort: 6.00 min. Review: 13.00 min. Cost: 2.50 USD.
Synthetic failure for demonstrating diagnosis. Illustrative observation; no execution or customer evidence.

receipt-followup · Extension robustness

Extend the receipt change with a second locale.

Acceptance criteria: Preserve the first requirement without duplicating the receipt pipeline.

RunOutcomeAcceptance / regressions / constraintsSafety / review
1passedpassed / passed / passedclear / accepted
Execution: 450000 ms. Time to acceptance: 1050000 ms. Corrective rescues: 0. Assistance effort: 0.00 min. Review: 10.00 min. Cost: 2.40 USD.
Synthetic outcome for demonstrating report calculations. Illustrative observation; no execution or customer evidence.
2passedpassed / passed / passedclear / accepted
Execution: 471000 ms. Time to acceptance: 1191000 ms. Corrective rescues: 0. Assistance effort: 0.00 min. Review: 12.00 min. Cost: 2.55 USD.
Synthetic outcome for demonstrating report calculations. Illustrative observation; no execution or customer evidence.
3passedpassed / passed / passedclear / accepted
Execution: 492000 ms. Time to acceptance: 1332000 ms. Corrective rescues: 0. Assistance effort: 0.00 min. Review: 14.00 min. Cost: 2.70 USD.
Synthetic outcome for demonstrating report calculations. Illustrative observation; no execution or customer evidence.

No universal autonomy certification, validated percentile or calibrated success prediction. Small and selected task sets limit generalisation. Missing evidence stays unknown. Imported reviewer and independence claims are supplied evidence, not attested facts. Production delivery is not established.