@claudemini/shit-cli 1.8.1 → 1.9.0

package/README.md

@@ -2,7 +2,7 @@
 
 **S**ession-based **H**ook **I**ntelligence **T**racker
 
-A zero-dependency memory system for human-AI coding sessions. Tracks what happened, classifies intent and risk, and provides structured data for code review automation.
+A memory system for human-AI coding sessions. Tracks what happened, classifies intent and risk, and provides structured data for code review automation.
 
 Supports **Claude Code**, **Gemini CLI**, **Cursor**, and **OpenCode**.
 
@@ -53,6 +53,7 @@ shit view <session-id>         # View semantic session report
 shit view <session-id> --json  # Include raw JSON data
 shit review [session-id]       # Run structured code review from session data
 shit review --json             # Machine-readable findings (structured schema)
+shit review                    # goatchain Agent-driven review
 shit review --recent=3 --md    # Aggregated Markdown report for PR comments
 shit review --strict --fail-on=medium  # CI gate by severity threshold
 shit explain <session-id>      # Human-friendly explanation of a session
@@ -64,6 +65,12 @@ shit explain <commit-sha>      # Explain a commit via its checkpoint
 - `--all` review all sessions in `.shit-logs`
 - `--min-severity=<info|low|medium|high|critical>` filter findings
 - `--fail-on=<info|low|medium|high|critical>` strict-mode failure threshold (default `high`)
+- `--agent-timeout-ms=<ms>` total timeout for the review agent (default `120000`)
+- `--allow-worktree-diff` allow worktree fallback diff when checkpoint commits are unavailable
+- `--agent-auto-approve=<bool>` auto approve agent tool calls (default `true`)
+- `--no-agent-auto-approve` disable auto approval for agent tool calls
+- `--agent` deprecated alias (accepted but ignored)
+- `--engine=<name>` deprecated alias (accepted but ignored)
 - `--strict` exit code `1` when findings reach `--fail-on`
 - `--json` output structured JSON
 - `--markdown` / `--md` output Markdown
@@ -112,7 +119,7 @@ shit summarize <session-id>    # Generate AI summary (requires API key)
 | `log <type>` | Log a hook event from stdin (called by hooks) |
 | `list` | List all sessions with type, intent, risk |
 | `view <id>` | View semantic session report |
-| `review [id]` | Run structured code review (single or multi-session) |
+| `review [id]` | Run agent-driven code review (goatchain, single or multi-session) |
 | `query` | Query session memory across sessions |
 | `explain <id>` | Human-friendly explanation of a session or commit |
 | `commit` | Create checkpoint on git commit |
@@ -139,7 +146,7 @@ Human <-> AI Agent (Claude Code, Gemini CLI, ...)
               |
          Memory System (.shit-logs/ + index.json)
               |
-         Code Review Bot / Human Queries
+         Agent Review (goatchain) / Human Queries
 ```
 
 1. **Ingestion** - Hooks fire on every agent event (tool use, prompts, session start/end). Events are appended to `events.jsonl`.
@@ -330,6 +337,7 @@ Set one of these environment variables to enable AI-powered session summaries:
 ```bash
 export OPENAI_API_KEY=sk-...      # Uses gpt-4o-mini by default
 export OPENAI_BASE_URL=https://api.openai.com/v1  # Optional: OpenAI-compatible base URL
+export AI_MODEL=gpt-5-mini         # Optional: universal model override (all providers)
 export ANTHROPIC_API_KEY=sk-...   # Uses claude-3-haiku by default
 ```
 
@@ -346,18 +354,29 @@ shit summarize <session-id>
 | `OPENAI_API_KEY` | Enable AI summaries via OpenAI |
 | `OPENAI_BASE_URL` | OpenAI-compatible base URL for summaries (default: `https://api.openai.com/v1`) |
 | `OPENAI_ENDPOINT` | Full OpenAI-compatible endpoint (overrides `OPENAI_BASE_URL`) |
+| `AI_MODEL` | Universal model override for summaries (applies to all providers) |
+| `OPENAI_MODEL` | OpenAI-only model override (takes precedence over `AI_MODEL` for OpenAI) |
 | `ANTHROPIC_API_KEY` | Enable AI summaries via Anthropic |
+| `ANTHROPIC_MODEL` | Anthropic-only model override (takes precedence over `AI_MODEL` for Anthropic) |
+| `SHIT_REVIEW_AGENT_TIMEOUT_MS` | Agent review timeout in milliseconds (default: `120000`) |
+| `SHIT_REVIEW_AUTO_APPROVE` | Auto approve agent tool calls (`true`/`false`, default: `true`) |
 | `SHIT_WEBHOOK_URL` | Webhook endpoint URL |
 | `SHIT_WEBHOOK_SECRET` | HMAC-SHA256 signing secret for webhooks |
 | `SHIT_WEBHOOK_AUTH_TOKEN` | Bearer token for webhook authentication |
 | `SHIT_WEBHOOK_EVENTS` | Comma-separated list of webhook events to subscribe to |
 
+`shit summarize` also accepts OpenAI-compatible responses where completion text is returned via `reasoning_content` instead of `content`.
+
 ## Security
 
 - Session logs are stored locally in `.shit-logs/` (added to `.gitignore` automatically)
 - Secrets (API keys, tokens, passwords) are automatically redacted when writing to shadow branches
 - Checkpoint data uses git plumbing commands — no impact on your working tree
 
+## Acknowledgements
+
+Special thanks to [goatchain](https://www.npmjs.com/package/goatchain) for powering the agent-driven review workflow in `shit review`.
+
 ## License
 
 MIT