output-format.md

JSON Output Format

When using --output-format json, the CLI outputs the full AnalysisResult object as pretty-printed JSON.

Top-level structure

{
  "metadata": { ... },
  "capability": { ... },
  "issues": [ ... ],
  "errors": [ ... ],
  "toolResults": [ ... ]
}

metadata

Execution metadata for the analysis run.

Field	Type	Description
startedAt	string	ISO 8601 timestamp
completedAt	string	ISO 8601 timestamp
durationMs	number	Total duration in milliseconds
repositoryRoot	string	Absolute path to the analyzed repository
executionMode	string	"standard", "strict", "auto-install", or "inspect"

capability

Tool availability report. Always present, even in non-inspect runs.

capability.ready[]

Tools that were available and ran (or could run in inspect mode).

Field	Type	Description
toolId	string	Tool identifier (e.g., "Ruff", "ESLint9")
displayName	string?	Human-readable name
version	string	Detected tool version
installation	string	"bundled", "global", or "codacy"
configSource	string?	"local" (native config file) or "codacy" (Codacy patterns)
configPath	string?	Path to native config file, when configSource is "local"
filesRouted	number?	Number of files routed to this tool
dependencies	array?	Resolved prerequisite statuses (name, version, installation)
runner	string	"local" or "container"

capability.unavailable[]

Tools that could not run.

Field	Type	Description
toolId	string	Tool identifier
displayName	string?	Human-readable name
reason	string	Why the tool is unavailable
remediation	string?	How to fix (e.g., install command)
availableVia	string?	"cloud", "container", or "manual-install"

issues[]

Array of issues found during analysis.

Field	Type	Description
filePath	string	Relative path from repository root
patternId	string	Rule identifier (e.g., "Ruff_E501")
toolId	string	Which tool found this issue
line	number	1-based start line
column	number	1-based start column
endLine	number?	1-based end line (if range)
endColumn	number?	1-based end column (if range)
message	string	Human-readable description
lineContent	string	Source code on the issue's line
multilineContent	string[]?	All source lines for multiline issues
category	string	Issue classification (see below)
severity	string	Issue severity (see below)
suggestion	string?	Suggested fix, if the tool provides one
confidence	number?	1 = low, 2 = medium, 3 = high

Severity values

See glossary.md for full severity definitions. The JSON field values map as follows:

Value	UI equivalent	Description
"Error"	Critical	Must fix
"High"	High	Should fix
"Warning"	Medium	Consider fixing
"Info"	Minor	Informational

Category values

See glossary.md for full category definitions.

"Security", "ErrorProne", "CodeStyle", "Compatibility", "Performance", "UnusedCode", "Complexity", "BestPractice", "Documentation"

errors[]

Pipeline errors (not code issues — tool crashes, parse failures, etc.).

Field	Type	Description
toolId	string	Which tool failed
phase	string	"requirementCheck", "requirementInstall", "toolInstall", "toolConfig", "toolInvoke", "outputParse"
kind	string	Machine-readable error kind (e.g., "MalformedOutput", "NonZeroExit")
message	string	Error description
level	string	"info", "warning", or "error"
filePath	string?	File where the error occurred, if applicable

toolResults[]

Per-tool execution summary.

Field	Type	Description
toolId	string	Tool identifier
status	string	"success", "partial", "failed", or "skipped"
issueCount	number	Issues found by this tool
errorCount	number	Errors from this tool
durationMs	number	Execution time in milliseconds
filesAnalyzed	number	Files this tool processed

Example: parsing results for agentic workflows

Filter critical and high-severity issues:

codacy-analysis analyze --output-format json | jq '.issues | map(select(.severity == "Error" or .severity == "High"))'

Get per-tool summary:

codacy-analysis analyze --output-format json | jq '.toolResults'

List unavailable tools with remediation:

codacy-analysis analyze --inspect --output-format json | jq '.capability.unavailable[] | {toolId, reason, remediation}'