SpecLock - AI Constraint Engine

Initialize SpecLock in the current project directory.

THE KEY TOOL. Returns the full structured context pack.

Set or update the project goal.

Add a non-negotiable constraint (SpecLock).

Remove (deactivate) a SpecLock by its ID.

Record an architectural or design decision.

Add a pinned note for reference.

Record deployment configuration facts.

Manually log a significant change.

Get recent file changes tracked by SpecLock.

Get the event log, optionally filtered by type.

Check if a proposed action conflicts with any active SpecLock. In hard mode, blocks above threshold.

Start a new session and get a full briefing.

End the current session and record what was accomplished.

Create a named git tag checkpoint for easy rollback.

Get current git repository status.

AI-powered lock suggestions based on project patterns.

Scan recent changes for constraint violations.

Health check with completeness score and multi-agent timeline.

Apply a pre-built constraint template (nextjs, react, express, supabase, stripe, security-hardened).

Violation report — how many times SpecLock blocked changes.

Audit staged files against active locks.

Verify the integrity of the HMAC audit chain.

Generate compliance reports (SOC 2, HIPAA, CSV).

Set enforcement mode: advisory (warn) or hard (block).

Override a lock with justification. Logged to audit trail.

Semantic pre-commit: analyzes code changes vs locks.

Show lock override history.

Evaluate policy-as-code rules against proposed actions.

Policy CRUD: list, add, remove policy rules.

Opt-in usage analytics summary.

Add SPECLOCK-GUARD header to lock specific files.

Auto-guard files related to lock keywords.

Add typed constraint (numerical/range/state/temporal).

Check proposed values against typed constraints.

List all typed constraints with current thresholds.

Update typed lock thresholds dynamically.

Compile natural language (PRDs, READMEs) into structured constraints via Gemini Flash.

Build/refresh code dependency graph from imports (JS/TS/Python).

Calculate blast radius — transitive dependents, impact %, depth.

Map active locks to actual code files via the dependency graph.

ALLOW/WARN/BLOCK verdict — combines semantic conflict + lock-file mapping + blast radius.

Diff-native review — parses actual diffs for interface breaks, protected symbols, dependency drift, schema changes.

Parse unified diff into structured changes — imports, exports, symbols, routes, schema detection.