cli.md

CLI Reference

Commands

lyingdocs analyze

Run a full documentation-code audit.

lyingdocs analyze --doc-path <docs> --code-path <code> [options]

Examples:

# Full analysis
lyingdocs analyze --doc-path docs/ --code-path . -o output/audit

# Choose Argus backend
lyingdocs analyze --doc-path docs/ --code-path . --argus-backend=local

# Different models for Hermes and Argus
lyingdocs analyze --doc-path docs/ --code-path . \
  --hermes-model gpt-5.4 \
  --argus-model gpt-5.4

# Resume interrupted analysis
lyingdocs analyze --doc-path docs/ --code-path . --resume

# Use an explicit config file
lyingdocs analyze --doc-path docs/ --code-path . --config myconfig.toml

# Generate GitHub issue drafts after analysis
lyingdocs analyze --doc-path docs/ --code-path . --gen-issue

lyingdocs init-ci

Generate a GitHub Actions workflow for LyingDocs CI integration.

lyingdocs init-ci [options]

Examples:

# Default: PR + tag triggers, local backend
lyingdocs init-ci --doc-path docs/ --code-path .

# Claude Code backend with OAuth token
lyingdocs init-ci --doc-path docs/ --backend claude_code --claude-oauth

# Only on tags + manual trigger, with approval gate
lyingdocs init-ci --doc-path docs/ --trigger tag,manual --approval

# Custom output path
lyingdocs init-ci --doc-path docs/ --trigger tag -o my-project/

See GitHub Actions Integration for the full setup guide.

lyingdocs version

Display the installed version.

lyingdocs version

---

Flags

analyze flags

Flag	Description
--doc-path	Path to documentation root
--code-path	Path to code root
-o, --output	Output directory for report artifacts
--config	Explicit path to a config file
--resume	Resume a previously interrupted analysis
--gen-issue	Generate a GitHub issue draft from findings
--hermes-model	Model name for Hermes
--hermes-base-url	API base URL for Hermes
--argus-backend	Argus backend: codex, claude_code, or local
--argus-model	Model name for Argus
--argus-base-url	API base URL for Argus
--argus-codex-provider	Provider flag passed to Codex CLI
--argus-codex-wire-api	Wire API for Codex backend (responses or chat)
--max-dispatches	Max number of Argus tasks per run
--max-iterations	Max Hermes agent loop iterations

init-ci flags

Flag	Description
--doc-path	Documentation root directory (default: docs/)
--code-path	Code repository root (default: .)
--backend	Argus backend: local, codex, or claude_code
--trigger	Comma-separated triggers: pr, tag, manual, schedule (default: pr,tag)
--branch	Target branch for PR trigger (default: main)
--cron	Cron expression for schedule trigger (default: 0 9 * * 1)
--approval	Add a manual approval step via GitHub Environments
--no-comment	Disable automatic PR comment with findings
--claude-oauth	Use Claude OAuth token instead of API key (Pro/Max subscription)
--gen-issue	Generate GitHub issue drafts from findings
--hermes-provider	LLM provider for Hermes: openai or anthropic (default: anthropic for claude_code, openai otherwise)
--hermes-model	Override Hermes LLM model
--argus-model	Override Argus LLM model
--action-ref	GitHub Action reference (default: KMing-L/lyingdocs@v1)
-o, --output	Output path (default: . — writes to .github/workflows/lyingdocs.yml)

---

Output artifacts

Each analysis run produces the following files in the output directory:

File	Description
Misalignment_Report.md	Structured report of all findings
findings.jsonl	Per-finding records (one JSON object per line)
doc_index.json	Documentation inventory built during audit
argus_task_NNN.txt	Raw output from each Argus task
workspace_state.json	Resume checkpoint
pipeline.log	Full execution log
issue.json	GitHub issue draft (only with --gen-issue)