addNextMessage extends the conversation with the assistant message and tool responses, applying three-zone compression at the soft (60%) and warning (

addOutputPathFlag registers --output/-o: write results to a UTF-8 file instead of stdout. "" or "-" keeps stdout; a real path is created lazily on the

apiKeyCmdForStep returns the api_key_cmd already configured for the provider the API-key step is editing, reading the same config entry loadExistingAP

applyLanguage appends instruction to all system-role messages in conv.

attemptCounts snapshots the per-file real HTTP attempt counts.

beginManualTokenReplace is the manual-tab equivalent of beginAPIKeyReplace.

buildBadge renders a compact "[category · severity]" tag for a finding. It returns an empty string when neither structured field is present, so text o

buildChangeFilesExceptGroup returns a formatted list of changed files excluding all group members.

buildDedupCommentsJSON renders the batch comments as a JSON list with stable c-N ids that the LLM groups by. Only fields the LLM needs to judge simila

buildGroupFilterCommentsJSON serializes comments with path info for group-level filtering.

buildMessageXML serializes msgs into the <message><content> form expected by the MEMORY_COMPRESSION_TASK prompt template.

buildSummaryCommentsList renders comments as a compact path-anchored markdown list suitable for embedding in the PROJECT_SUMMARY_TASK prompt. Format:

Build the ``[category · severity]`` badge (middot U+00B7). Byte-matches the CLI's buildBadge degeneration so review output is consistent acro

canonicalRemote reduces a git remote URL to a stable, credential-free identity string for hashing into repository.identity_sha256, so the same reposit

checkAPIKeyRequirement decides whether a provider selection may be saved with no api_key. It mirrors the resolver's precedence (static api_key -> api_

checkMetricErr intentionally ignores metric registration errors: telemetry is best-effort and must not interrupt the main flow.

classifyBoundaryError classifies an error that only surfaced after the SDK retry loop finished, and reports whether it recognized it at all. The fals

classifyItemError maps a subtask error to a stable item failure class and a safe, generic reason. It never returns the raw error text (which may embed

classifyMainLoopStop maps a non-error, non-completed main-loop stop to an item failure class and a safe reason. Only the configured max-tool-request b

classifyStreamError classifies a failure that surfaced while consuming an already-established SSE stream. Unlike classifyBoundaryError it always retur

colorf formats then colorizes, so call sites need not nest fmt.Sprintf.

commitFile writes content to name and commits it.

copyMessages creates a shallow copy of a message slice.

countLines returns the number of lines in content. A file without a trailing newline still counts its final line. Empty input → 0.

countReviewable counts diffs that will survive all filters and are not pure deletions.

cursorOnDeletableModel reports whether the model-step cursor is on a row that can be deleted (not on "Enter custom model name...").

customAPIKeyForSave reports the API key to persist and whether the user edited the field (vs left the masked placeholder untouched).

decodeSinglePreviewJSON asserts that s is exactly one JSON value followed only by the encoder's trailing newline. Automation consuming --format json r

errBedrockNotConfigurable explains why the two url+token strategies reject the bedrock protocol. Both describe a single HTTP endpoint and carry no pla

estimateDiffCost projects token usage for reviewing the given diffs. It is the diff-path analogue of scan.estimateCost and is used for the pre-review

estimateCost projects token usage for reviewing the given items under the supplied scan template. planEnabled / dedupEnabled / summaryEnabled reflect

extFromPath returns the file extension with leading dot, lowercased.

extractCodeBlock extracts the content of the first fenced code block from text. Returns empty string if no code block is found.

Render ``content`` as a fenced code block with a safe fence.

fileReadRef picks the ref file_read resolves paths against. A sealed input replaces the ref the user typed with the commit that ref resolved to at ad

filterByPaths keeps only entries whose path equals a user-supplied path (for exact files) or lies under it (for directories).

filterScanItems drops items that should not be reviewed under the standard reviewability rules (binary, extension allowlist, user include/exclude, def

Return the newest note carrying the summary marker (or per-run tag). ``tag`` selects the matching needle: when provided, match this run's per