← Prosights AEI on Prosights.ai. This line is not part of the report: the website adds it to the sample. Project Larch, its 3 repositories, the assessor and every result are made up.

Pre-deal assessment: Project Larch

How the team behind Project Larch controls its work with coding agents, graded with the Prosights Agentic Engineering Index (AEI). The target ran AEI Scan on its own machines, gave its own answers and uploaded 3 results for 3 repositories. This site computed every score again from the check results. Each figure below says how well it is evidenced.

For the investment committee

ResultOverall score 82.5 to 86.4, median 85.9, of 100, over 3 results
Level1, Written
EvidenceWeakest grade of any graded check: Declared. Share of the score at each grade: Declared 38.5%, Scanned 36.8%, Witnessed 22.8%, Verified by assessor 2%
Coverage160 of 291 applicable checks are graded (55%)
Scope, as the target declared itThe target states 9 repositories with coding-agent use and uploaded results for 3. Nobody checked this statement.
Claim gap5 checks where the target said yes and the scan found nothing
Open checks by severityCritical 2, High 3, Medium 2, Low 11
DatesRequested 2026-11-14; agreed 2026-11-14; computed 2026-11-14; the request ends 2026-12-31

In scope, in the target's words: "The 3 repositories of the payments product: payments-api, payments-web and ledger-service." Out of scope, in the target's words: "6 repositories of internal tools. 2 of them had coding-agent work this year."

The 5 most serious findings

SeverityCheckThe practice the check asks forResultEvidence grade
Critical9.1, level 3The agent's sandbox is on.open in 2 of 3 results: fail, claimedDeclared
Critical9.2, level 3The agent's network access is limited to a list.open in 3 of 3 results: claimed, failDeclared
High1.4, level 3A hook runs when a prompt is submitted to the agent.open in 2 of 3 results: fail, claimedDeclared
High6.3, level 3A hook runs when the agent ends its turn.open in 1 of 3 results: failWitnessed
High15.11, level 3No command can run outside the sandbox.open in 3 of 3 results: claimed, failDeclared

The 16 dimensions

#DimensionScoreLowest levelWeakest grade
1Requirements and recording63 to 100, median 631, WrittenDeclared
2Decisions50 to 100, median 1001, WrittenDeclared
3Expected behavior and designnot ratednot ratednone
4Agent briefing and model policy70 to 100, median 701, WrittenDeclared
5Build70 to 100, median 1001, WrittenDeclared
6Test and evidence50 to 100, median 1002, CheckedScanned
7Review and mergenot ratednot ratedDeclared
8Releasenot ratednot ratedDeclared
9Safety controls63 to 81, median 811, WrittenDeclared
10Memory and continuity1001, WrittenDeclared
11Measurementnot ratednot ratedDeclared
12Governance1003, EnforcedDeclared
13Token and cost64 to 100, median 811, WrittenDeclared
14Harness depth63 to 100, median 811, WrittenDeclared
15Process guardsnot ratednot ratedDeclared
16Model conductnot ratednot ratedDeclared

Where the results differ, a score reads "lowest to highest, median". 3 results from 3 repositories: no result is an average of the others.

Findings by severity

A finding is a check that is open: it failed, the target declined it, or the target claimed it with no evidence. The rule for each severity is written here, and it is the whole rule:

SeverityCheckThe practice the check asks forWhat the check readsResult and reasonEvidence grade
Critical9.1, level 3The agent's sandbox is on.The sandbox settingpayments-api (Claude Code): fail; the setting is off / payments-web (Cursor): claimed; no documented location on Cursor; the answer was yesDeclared
Critical9.2, level 3The agent's network access is limited to a list.The setting that limits network accessledger-service (Codex CLI): claimed; no documented location on Codex CLI; the answer was yes / payments-api (Claude Code): fail; the setting is not set / payments-web (Cursor): claimed; no documented location on Cursor; the answer was yesDeclared
High1.4, level 3A hook runs when a prompt is submitted to the agent.A hook on the prompt-submit event (Claude Code and Codex CLI UserPromptSubmit; VS Code userPromptSubmitted; on other platforms, shown to a person)payments-api (Claude Code): fail; no hook registered on UserPromptSubmit / payments-web (Cursor): claimed; no documented location on Cursor: no hook event for this is documented; the answer was yesDeclared
High6.3, level 3A hook runs when the agent ends its turn.A hook on the end of the agent's turn (Claude Code and Codex CLI Stop; Cursor stop; VS Code agentStop; on other platforms, shown to a person)payments-api (Claude Code): fail; no hook registered on StopWitnessed
High15.11, level 3No command can run outside the sandbox.The setting that lets a command run outside the sandboxledger-service (Codex CLI): claimed; no documented location on Codex CLI; the answer was yes / payments-api (Claude Code): fail; the setting is not set / payments-web (Cursor): claimed; no documented location on Cursor; the answer was yesDeclared
Medium2.2, level 2A check reads the decision records at commit.The check that reads decision records at commitpayments-api (Claude Code): fail; assessor, 2026-11-12: Asked for; the team had no artifact to show.Verified by assessor
Medium9.5, level 3When a hook itself fails, the action is refused.What happens when the hook itself failsledger-service (Codex CLI): claimed; no documented location on Codex CLI: no setting for a failing hook is documented; the answer was yesDeclared
Low16.4, level 2A hook runs when a tool call is denied.A hook on a denied tool call (Claude Code PermissionDenied; Cursor postToolUseFailure; on other platforms, shown to a person)ledger-service (Codex CLI): claimed; no documented location on Codex CLI: no hook event for this is documented; the answer was yesDeclared
Low4.5, level 3The agent's model is named with a version number.The model named in settings and in agent definitionsledger-service (Codex CLI): claimed; no documented location on Codex CLI; the answer was yes / payments-web (Cursor): claimed; no documented location on Cursor; the answer was yesDeclared
Low4.6, level 3A hook runs when an agent starts another agent.A hook on the tool that starts another agent (Claude Code PreToolUse on Agent; on other platforms, shown to a person)ledger-service (Codex CLI): claimed; no documented location on Codex CLI: no hook event for this is documented; the answer was yes / payments-web (Cursor): claimed; no documented location on Cursor: no hook event for this is documented; the answer was yesDeclared
Low5.2, level 3A hook runs before the agent's shell commands.A hook before shell commands (Claude Code PreToolUse on Bash; Cursor beforeShellExecution; on other platforms, shown to a person)ledger-service (Codex CLI): claimed; no documented location on Codex CLI: no hook event for this is documented; the answer was yesDeclared
Low13.3, level 3A threshold is set at which the conversation is compacted.The setting for compacting the conversationpayments-web (Cursor): claimed; no documented location on Cursor; the answer was yesDeclared
Low13.4, level 3The reasoning effort is set in settings or in an agent definition.The reasoning-effort settingpayments-web (Cursor): claimed; no documented location on Cursor; the answer was yesDeclared
Low13.6, level 3A prompt cache lifetime is set, and caching is on.The prompt-cache settingsledger-service (Codex CLI): claimed; no documented location on Codex CLI; the answer was yes / payments-web (Cursor): claimed; no documented location on Cursor; the answer was yesDeclared
Low14.1, level 3A hook runs before the agent writes or edits a file.A hook that runs before the agent writes or edits a file (Claude Code PreToolUse on Write or Edit)ledger-service (Codex CLI): claimed; no documented location on Codex CLI: no hook event for this is documented; the answer was yes / payments-web (Cursor): claimed; no documented location on Cursor: no hook event for this is documented; the answer was yesDeclared
Low14.3, level 3An agent definition limits the tools its agent may use.The tool limits in agent definitionsledger-service (Codex CLI): claimed; no documented location on Codex CLI; the answer was yesDeclared
Low11.1, level 4Session records of the last 30 days are kept.The count and dates of the agent's session records, never their text, and only when you turn the count onpayments-web (Cursor): claimed; no documented location on Cursor; the answer was yesDeclared
Low13.12, level 4Usage telemetry is on.The usage-telemetry settingledger-service (Codex CLI): claimed; no documented location on Codex CLI; the answer was yes / payments-web (Cursor): claimed; no documented location on Cursor; the answer was yesDeclared

Questions for management

1 question for each check that has no result, that the target declined, or that rests on the target's own statement: 81 questions. Each asks to be shown what would pass the check.

CheckWhy it is askedAsk to be shown
1.1, level 1rests on the target's own statementWork items are written down before an agent starts on them. The check reads: Your answer to the survey question "Are work items written down before an agent starts on them?".
1.2, level 2not answered or not readableA check refuses a change that is tied to no work item. The check reads: The check that refuses a change tied to no item.
1.3, level 2not answered or not readableA script regenerates the list of open work. The check reads: A list of open work that a script regenerates.
1.4, level 3rests on the target's own statementA hook runs when a prompt is submitted to the agent. The check reads: A hook on the prompt-submit event (Claude Code and Codex CLI UserPromptSubmit; VS Code userPromptSubmitted; on other platforms, shown to a person).
2.2, level 2not answered or not readableA check reads the decision records at commit. The check reads: The check that reads decision records at commit.
2.3, level 1rests on the target's own statementIt is written which choices a person makes and which an agent may make. The check reads: Your answer to the survey question "Is it written which choices a person makes and which an agent may make?".
3.1, level 1not answered or not readableThe expected behaviors of a change are written before the build. The check reads: For 1 recent change: the expected behaviors written before the build.
3.2, level 2not answered or not readableA check ties each expected behavior to a test. The check reads: The check that ties each expected behavior to a test.
4.3, level 1rests on the target's own statementRule files apply only to some paths. The check reads: Rule files that apply only to some paths.
4.5, level 3rests on the target's own statementThe agent's model is named with a version number. The check reads: The model named in settings and in agent definitions.
4.6, level 3rests on the target's own statementA hook runs when an agent starts another agent. The check reads: A hook on the tool that starts another agent (Claude Code PreToolUse on Agent; on other platforms, shown to a person).
5.1, level 1rests on the target's own statementEach agent works in its own checkout or environment. The check reads: Your answer to the survey question "Does each agent work in its own checkout or environment?".
5.2, level 3rests on the target's own statementA hook runs before the agent's shell commands. The check reads: A hook before shell commands (Claude Code PreToolUse on Bash; Cursor beforeShellExecution; on other platforms, shown to a person).
5.3, level 2not answered or not readableA limit stops a build or test run that hangs. The check reads: The limit that stops a build or test run that hangs.
5.4, level 1rests on the target's own statementThe size of 1 agent task has a written limit. The check reads: Your answer to the survey question "Is there a written limit on the size of 1 agent task?".
6.4, level 4not answered or not readableThe test results of each build are kept as files. The check reads: Test results kept as files for each build.
6.5, level 4not answered or not readableA test set for each crucial function runs at every release. The check reads: For each function you name as crucial: a test set run at every release.
6.6, level 2not answered or not readableA gate fails on new warnings in changed files. The check reads: The gate that fails on new warnings in changed files.
7.1, level 1rests on the target's own statementA reviewer other than its author reviews every agent change before it merges: a person, or a model run as a separate reviewer. The check reads: Your answer to the survey question "Is every agent change reviewed before it merges by a reviewer other than its author: a person, or a model run as a separate reviewer?".
7.3, level 2not answered or not readableAn automated review runs on every change before it merges and cannot be skipped. The check reads: An automated review that runs on every change before it merges and cannot be skipped.
7.4, level 2not answered or not readableA merge step builds and checks after each merge and stops on failure. The check reads: The merge step that builds and checks after each merge and stops on failure.
7.5, level 1rests on the target's own statementIt is written who may start or assign an agent. The check reads: Your answer to the survey question "Is it written who may start or assign an agent?".
7.6, level 3not answered or not readableA written rule says which changes need a person's approval before they merge or ship, by risk. The check reads: The written rule that says which changes need a person's approval before they merge or ship, by risk.
7.7, level 4not answered or not readableAutomated approvals are recorded, and a person reviews a sample of them each quarter. The check reads: The record of automated approvals, and the sample of them a person reviewed in the last quarter.
8.1, level 1rests on the target's own statementA person confirms each release. The check reads: Your answer to the survey question "Does a person confirm each release?".
8.2, level 2not answered or not readableA release gate refuses when a required item is missing. The check reads: The release gate that refuses when a required item is missing.
8.3, level 4not answered or not readableThe gate result of each release is recorded. The check reads: The record of each release's gate result.
9.1, level 3rests on the target's own statementThe agent's sandbox is on. The check reads: The sandbox setting.
9.2, level 3rests on the target's own statementThe agent's network access is limited to a list. The check reads: The setting that limits network access.
9.5, level 3rests on the target's own statementWhen a hook itself fails, the action is refused. The check reads: What happens when the hook itself fails.
9.6, level 2not answered or not readableA secret scan runs before code leaves the machine. The check reads: The secret scan that runs before code leaves the machine.
9.8, level 4not answered or not readableA list marks each of the team's rules as enforced, checked or written only. The check reads: A list of the team's rules marked enforced, checked or written only.
10.1, level 1rests on the target's own statementThe agent keeps memory files for this project. The check reads: How many files the agent's memory folder holds, never their contents.
10.2, level 2not answered or not readableAn end-of-session step refuses to finish with failing tests or unpushed work. The check reads: The end-of-session step that refuses to finish with failing tests or unpushed work.
10.3, level 1rests on the target's own statementThe start-of-session procedure is written. The check reads: Your answer to the survey question "Is the start-of-session procedure written?".
10.4, level 1rests on the target's own statementEach agent writes down what it learned from its work, in a place the next agent reads. The check reads: Your answer to the survey question "Does each agent write down what it learned from its work, in a place the next agent reads?".
10.5, level 2not answered or not readableA mechanism refuses a change whose agent recorded no lesson entry. The check reads: The mechanism that refuses a change whose agent recorded no lesson entry.
10.6, level 2not answered or not readableFor each of the last 5 agent changes, a record shows that the agent read the carried-forward lessons before its work. The check reads: For the last 5 agent changes, the record that the agent read the carried-forward lessons before its work.
11.1, level 4rests on the target's own statementSession records of the last 30 days are kept. The check reads: The count and dates of the agent's session records, never their text, and only when you turn the count on.
11.2, level 4not answered or not readableThe merge rate and the revert rate of agent changes are measured against the team's own baseline. The check reads: Merge rate and revert rate of agent changes against the team's own baseline.
11.3, level 4not answered or not readableUsage or cost is recorded per agent or task. The check reads: Usage or cost recorded per agent or task.
11.4, level 4not answered or not readableA quality trend is kept over time: warnings, complexity or failures. The check reads: A quality trend: warnings, complexity or failures over time.
11.5, level 1rests on the target's own statementAgent work is measured by a named outcome. The check reads: Your answer to the survey question "Do you measure agent work by a named outcome?".
12.1, level 1rests on the target's own statementA written statement says which AI tools are used, on which data, and who approves. The check reads: Your answer to the survey question "Is there a written statement of which AI tools are used, on which data, and who approves?".
12.3, level 1not answered or not readableThe rules have stable identifiers and a record of why each exists. The check reads: Rules with stable identifiers and a record of why each exists.
12.4, level 1rests on the target's own statementAgent use is covered by the organization's AI risk process. The check reads: Your answer to the survey question "Is agent use covered by your organization's AI risk process?".
13.1, level 1rests on the target's own statementIt is written which model and which reasoning effort each kind of task uses. The check reads: Your answer to the survey question "Is it written which model and which reasoning effort each kind of task uses?".
13.2, level 1not answered or not readableEvery instruction file has fewer than 200 lines. The check reads: The number of lines in each instruction file, never the lines.
13.3, level 3rests on the target's own statementA threshold is set at which the conversation is compacted. The check reads: The setting for compacting the conversation.
13.4, level 3rests on the target's own statementThe reasoning effort is set in settings or in an agent definition. The check reads: The reasoning-effort setting.
13.6, level 3rests on the target's own statementA prompt cache lifetime is set, and caching is on. The check reads: The prompt-cache settings.
13.7, level 4not answered or not readableA report separates input tokens read from the cache from input tokens sent uncached. The check reads: A report that separates input tokens read from the cache from input tokens sent uncached.
13.8, level 2not answered or not readableA mechanism shortens long command output before the agent reads it. The check reads: A mechanism that shortens long command output before the agent reads it.
13.9, level 1rests on the target's own statementIt is written that long or verbose work goes to its own agent, and that agents wait without polling. The check reads: Your answer to the survey question "Is it written that long or verbose work goes to its own agent, and that agents wait without polling?".
13.10, level 3not answered or not readableA spend or usage limit is set at the provider. The check reads: A spend or usage limit set at the provider, for the workspace or for each person.
13.11, level 4not answered or not readableTokens or cost per model are reviewed over time, and a change was made because of a review. The check reads: A review of tokens or cost per model over time, and 1 change made because of it.
13.12, level 4rests on the target's own statementUsage telemetry is on. The check reads: The usage-telemetry setting.
14.1, level 3rests on the target's own statementA hook runs before the agent writes or edits a file. The check reads: A hook that runs before the agent writes or edits a file (Claude Code PreToolUse on Write or Edit).
14.3, level 3rests on the target's own statementAn agent definition limits the tools its agent may use. The check reads: The tool limits in agent definitions.
14.5, level 2not answered or not readableA command shows the state of each background or unattended run. The check reads: The command that shows the state of each background or unattended run.
14.6, level 3not answered or not readableAn agent runs commands on another machine only from an allow list. The check reads: The channel through which an agent runs a command on another machine.
14.7, level 2not answered or not readableThe team's own hooks and guard scripts have tests. The check reads: The tests of the team's own hooks and guard scripts.
15.1, level 3not answered or not readableEvery delete goes through 1 script, and a hook refuses a delete made any other way. The check reads: The script every delete goes through, and the hook that refuses a delete made any other way.
15.2, level 3not answered or not readableA hook refuses a build or test command run outside the team's wrapper script. The check reads: The hook that refuses a build or test command run outside the team's wrapper script.
15.3, level 2not answered or not readableA check keeps credentials and real messages out of test fixtures. The check reads: The check that keeps credentials and real messages out of test fixtures.
15.4, level 1rests on the target's own statementIt is written that a new integration or tool starts with no access until it is listed. The check reads: Your answer to the survey question "Is it written that a new integration or tool starts with no access until it is listed?".
15.5, level 2not answered or not readableA mechanism stops a test from reaching a live account unless a person turns that on. The check reads: What stops a test from reaching a live account unless a person turns that on.
15.6, level 2not answered or not readableA commit check refuses personal identifiers and authorship marks. The check reads: The commit check that refuses personal identifiers and authorship marks.
15.7, level 2not answered or not readableA set of injection attempts is run against a model before it is given a channel. The check reads: The set of injection attempts run against a model before it is given a channel, and the step that runs it.
15.8, level 1rests on the target's own statementIt is written that a change touches only what its own work item needs. The check reads: Your answer to the survey question "Is it written that a change touches only what its own work item needs?".
15.9, level 2not answered or not readableA guard has a single switch, and its off state is tested. The check reads: For 1 guard: its single switch, and the test of the off state.
15.10, level 3not answered or not readableA script removes an agent's workspace and refuses when its work is not merged. The check reads: The script that removes an agent's workspace and refuses when its work is not merged.
15.11, level 3rests on the target's own statementNo command can run outside the sandbox. The check reads: The setting that lets a command run outside the sandbox.
16.1, level 1rests on the target's own statementA written record shows how each model in use has failed in the team's own work. The check reads: Your answer to the survey question "Is there a written record of how each model you use has failed in your own work?".
16.2, level 1rests on the target's own statementEvery brief to an agent states what done means. The check reads: Your answer to the survey question "Does every brief to an agent state what done means?".
16.3, level 2not answered or not readableSomeone other than the agent checks the claims in its reports. The check reads: For the last 2 agent reports: who other than the agent checked the claims, and what they checked.
16.4, level 2rests on the target's own statementA hook runs when a tool call is denied. The check reads: A hook on a denied tool call (Claude Code PermissionDenied; Cursor postToolUseFailure; on other platforms, shown to a person).
16.5, level 2not answered or not readableRefused calls are logged, each with the model that made the call. The check reads: The log of refused calls, each with the model that made the call.
16.6, level 3not answered or not readableA mechanism refuses the end of a turn while tasks are open. The check reads: The mechanism that refuses the end of a turn while tasks are open, and 1 refusal it made.
16.8, level 4not answered or not readableRule breaks, early stops and false "done" reports are counted per model, and the counts are read on a schedule. The check reads: Counts per model for 2 periods (rule breaks, early stops, false "done" reports) and the schedule on which they are read.
16.9, level 4not answered or not readableA new model version was measured before it was adopted. The check reads: The measurement made before the last new model version was adopted, and the line of the model policy it changed or confirmed.

Remediation plan

Every open check, the lowest level first, and within a level by the kind of work. The 3 kinds of work:

No estimate of hours or cost is given: none was measured. The target's own report from AEI Scan names the exact setting for each scan check.

LevelKind of workCheckWhat to put in placeWhereOpen in
2Build a mechanism and show it2.2Add a check that reads the decision records at commit.shown to a person1 of 3
2Build a mechanism and show it16.4Register a hook that runs when a tool call is denied.shown to a person1 of 3
3Add a setting or a file9.1Turn the agent's sandbox on..claude/settings.json, .claude/settings.local.json, ~/.claude/settings.json2 of 3
3Add a setting or a file9.2Limit the agent's network access to a list..claude/settings.json, .claude/settings.local.json, ~/.claude/settings.json3 of 3
3Add a setting or a file1.4Register a hook that runs when a prompt is submitted to the agent..claude/settings.json, .claude/settings.local.json, ~/.claude/settings.json2 of 3
3Add a setting or a file6.3Register a hook that runs a check when the agent ends its turn..claude/settings.json, .claude/settings.local.json, ~/.claude/settings.json1 of 3
3Add a setting or a file15.11Set the sandbox so that no command can run outside it..claude/settings.json, .claude/settings.local.json, ~/.claude/settings.json3 of 3
3Build a mechanism and show it9.5Set every hook so that the action is refused when the hook itself fails.shown to a person1 of 3
3Build a mechanism and show it4.5Name the agent's model with a version number, not by an alias.shown to a person2 of 3
3Build a mechanism and show it4.6Register a hook on the tool that starts another agent.shown to a person2 of 3
3Build a mechanism and show it5.2Register a hook that runs before the agent's shell commands.shown to a person1 of 3
3Build a mechanism and show it13.3Set a threshold at which the agent's conversation is compacted.shown to a person1 of 3
3Build a mechanism and show it13.4Set the agent's reasoning effort in settings or in an agent definition.shown to a person1 of 3
3Build a mechanism and show it13.6Set a prompt cache lifetime, and leave caching on.shown to a person2 of 3
3Build a mechanism and show it14.1Register a hook that runs before the agent writes or edits a file.shown to a person2 of 3
3Build a mechanism and show it14.3Limit the tools of at least 1 agent in its agent definition.shown to a person1 of 3
4Build a mechanism and show it11.1Keep the agent's session records for at least 30 days.shown to a person1 of 3
4Build a mechanism and show it13.12Turn usage telemetry on for the agent.shown to a person2 of 3

How well each figure is evidenced

GradeWhat it meansGraded checksShare of the score
DeclaredThe target said so: a survey answer, or the target's own statement that an artifact exists. Nobody outside the target saw it.7938.5%
ScannedAEI Scan read it on the target's machine, in a run that nobody outside the target watched.4336.8%
WitnessedAEI Scan read it in a run that an assessor watched. The assessor recorded the date and what was watched.3522.8%
Verified by assessorAn assessor was shown the artifact and graded the check. The assessor recorded the date and what was shown.32%

A check that failed carries no points, so a grade can hold many checks and a small share. A witnessed run shows the repository at the moment of the run, and a team that is watched may behave differently. An unobserved run can be staged.

The results, 1 for each repository and coding agent

RepositoryCoding agentScoreLevelCoverageClaim gapWeakest gradeRun
ledger-serviceCodex CLI86.41, Written54%0DeclaredNot watched
payments-apiClaude Code82.51, Written58%5DeclaredWitnessed 2026-11-12: A screen share of the whole run, started from a fresh clone of the repository.
payments-webCursor85.91, Written54%0DeclaredNot watched

Appendix: integrity of each file

RepositoryCoding agentSHA-256 of the uploaded fileToolCatalogRun (UTC)Uploaded (UTC)
ledger-serviceCodex CLIe595e1595b89d11b6253cad3bbb10be79b69a99a785c7746960a7bb9773f3d4dAEI Scan 0.1.02026-11-162026-11-16 19:422026-11-14 12:00
payments-apiClaude Code056178bc7584389c92c47a8249787bec1787a6c7c67f897a90644dc03ec59fb2AEI Scan 0.1.02026-11-162026-11-16 19:422026-11-14 12:00
payments-webCursor28eef9f2aa1fd0fe9e7e42929142625fd100add4409ef6b7a3db11ab23957ac8AEI Scan 0.1.02026-11-162026-11-16 19:422026-11-14 12:00

The SHA-256 shows that the stored file is the file AEI Scan printed that value for. It does not show who ran the scan or on which machine: the file carries no signature, and the results file does not record the operating system. The names of the repositories are the folder names the uploads gave.

Appendix: said but not found, and in place but not known

Said but not found: the target answered yes and the scan found nothing.

In place but not known: the scan found it and the target answered no.

Appendix: token and cost (dimension 13)

CheckThe practice the check asks forledger-service, Codex CLIpayments-api, Claude Codepayments-web, Cursor
13.1, level 1It is written which model and which reasoning effort each kind of task uses.pass (Declared)pass (Declared)pass (Declared)
13.2, level 1Every instruction file has fewer than 200 lines.unknownpass (Witnessed)unknown
13.3, level 3A threshold is set at which the conversation is compacted.pass (Scanned)pass (Witnessed)claimed (Declared)
13.4, level 3The reasoning effort is set in settings or in an agent definition.pass (Scanned)pass (Witnessed)claimed (Declared)
13.5, level 3Agents that another agent starts are given their own model.pass (Scanned)pass (Witnessed)pass (Scanned)
13.6, level 3A prompt cache lifetime is set, and caching is on.claimed (Declared)pass (Witnessed)claimed (Declared)
13.7, level 4A report separates input tokens read from the cache from input tokens sent uncached.unknownunknownunknown
13.8, level 2A mechanism shortens long command output before the agent reads it.unknownunknownunknown
13.9, level 1It is written that long or verbose work goes to its own agent, and that agents wait without polling.pass (Declared)pass (Declared)pass (Declared)
13.10, level 3A spend or usage limit is set at the provider.unknownunknownunknown
13.11, level 4Tokens or cost per model are reviewed over time, and a change was made because of a review.unknownunknownunknown
13.12, level 4Usage telemetry is on.claimed (Declared)pass (Witnessed)claimed (Declared)

Appendix: model conduct (dimension 16)

CheckThe practice the check asks forledger-service, Codex CLIpayments-api, Claude Codepayments-web, Cursor
16.1, level 1A written record shows how each model in use has failed in the team's own work.pass (Declared)pass (Declared)pass (Declared)
16.2, level 1Every brief to an agent states what done means.pass (Declared)pass (Declared)pass (Declared)
16.3, level 2Someone other than the agent checks the claims in its reports.unknownunknownunknown
16.4, level 2A hook runs when a tool call is denied.claimed (Declared)pass (Witnessed)pass (Scanned)
16.5, level 2Refused calls are logged, each with the model that made the call.unknownunknownunknown
16.6, level 3A mechanism refuses the end of a turn while tasks are open.unknownunknownunknown
16.7, level 3A hook runs when a started agent ends its work or a task is marked complete.pass (Scanned)pass (Witnessed)pass (Scanned)
16.8, level 4Rule breaks, early stops and false "done" reports are counted per model, and the counts are read on a schedule.unknownunknownunknown
16.9, level 4A new model version was measured before it was adopted.unknownunknownunknown

The counts below are the target's own, for each model its agents run on: how often the model broke a rule, stopped early or asked for a permission it had, reported done when it was not, invented a reason, or did work nobody asked for. AEI Scan counted none of them and they change no grade. Their evidence grade is Declared.

RepositoryModelPeriodRules brokenStopped early or asked for a permission it hadReported done when it was notInvented a reasonDid work nobody asked for
payments-apiModel A2026-0996324
payments-apiModel A2026-1047125
payments-apiModel B2026-10212

Appendix: how the results compare

RepositoryCoding agentScoreA default installation of this coding agentAmong all shared results
ledger-serviceCodex CLI86.4passes 0 of 35 scan checks; this result passes 23no comparison: 0 shared results, fewer than 11
payments-apiClaude Code82.5passes 0 of 35 scan checks; this result passes 30no comparison: 0 shared results, fewer than 11
payments-webCursor85.9passes 0 of 35 scan checks; this result passes 22no comparison: 0 shared results, fewer than 11

The default installation is the coding agent as installed, with nothing configured (catalog of 2026-11-16). The shared results are those whose owners chose to share them; they are not a sample of all teams, and a figure is shown only for a group of at least 11.

What this report does not show