ecto-constraint-debug
# ecto-constraint-debug This skill provides systematic debugging methods for Ecto constraint violations in Elixir Phoenix applications. Use it when encountering `Ecto.ConstraintError` or changeset errors related to unique constraints, foreign key constraints, or check constraints by parsing error messages, locating constraint definitions in migrations, tracing all insert/update code paths, and identifying root causes like race conditions, data validation gaps, or concurrent request issues.
git clone --depth 1 https://github.com/oliver-kriska/claude-elixir-phoenix /tmp/ecto-constraint-debug && cp -r /tmp/ecto-constraint-debug/plugins/elixir-phoenix/skills/ecto-constraint-debug ~/.claude/skills/ecto-constraint-debugSKILL.md
# Ecto Constraint Debugging
> **Ash projects**: Ash surfaces DB constraints through its own error DSL. Use the `ash-framework` skill — `mix usage_rules.search_docs "constraint" -p ash_postgres`.
Systematic approach to diagnosing constraint violations. Load when you see `Ecto.ConstraintError`, `unique_constraint`, `foreign_key_constraint`, or constraint-related changeset errors.
## Iron Laws
1. **READ THE CONSTRAINT NAME** — The constraint name (e.g., `links_url_index`) tells you exactly which index/constraint failed. Parse it from the error message first
2. **CHECK MIGRATION BEFORE CODE** — Verify the constraint definition in `priv/repo/migrations/` matches what the schema expects
3. **TRACE ALL INSERT PATHS** — Find every code path that inserts into the constrained table. The bug is often in a path you didn't consider
4. **RACE CONDITION UNTIL PROVEN OTHERWISE** — If validation passes but constraint fails, assume concurrent inserts until you prove a single-request cause
## Step-by-Step Debugging
### Step 1: Parse the Error
Extract from the error message:
- **Constraint name** (e.g., `users_email_index`)
- **Table name** (e.g., `users`)
- **Operation** (insert, update, or delete)
- **Conflicting values** (if available in logs)
### Step 2: Find the Migration
Use Grep to search for the constraint name in `priv/repo/migrations/`. Also check for `create unique_index`, `create index`, `add constraint`.
Verify: Does the migration constraint match the schema's `unique_constraint/3` or `foreign_key_constraint/3` call?
### Step 3: Find the Schema
Use Grep to find constraint handling in changesets (`unique_constraint`, `foreign_key_constraint`, `check_constraint`) in `lib/`.
### Step 4: Trace Insert Paths
Find ALL callers that insert/update this schema:
Use Grep to find all insert/update paths (`Repo.insert`, `Repo.update`, `Repo.insert_all`, `cast_assoc`) in `lib/`.
### Step 5: Identify the Cause
| Symptom | Likely Cause | Fix Pattern |
|---------|-------------|-------------|
| Same user triggers twice | Race condition (double-click, retry) | Upsert with `on_conflict` |
| Multiple parents share child | `cast_assoc` doesn't dedup across changesets | Dedup before building changesets |
| Concurrent API requests | Missing transaction isolation | Wrap in `Repo.transaction` or use upsert |
| Migration added constraint to existing data | Data violates new constraint | Backfill or clean data first |
### Step 6: Apply Fix
See `${CLAUDE_SKILL_DIR}/references/constraint-patterns.md` for detailed fix patterns.
## Quick Fixes by Constraint Type
**Unique violation** → Upsert: `Repo.insert(changeset, on_conflict: :replace_all, conflict_target: [:field])`
**Foreign key violation** → Check: Does the referenced record exist? Was it deleted concurrently?
**Check constraint** → Validate: Does the value satisfy the constraint condition?
## References
- `${CLAUDE_SKILL_DIR}/references/constraint-patterns.md` - Detailed patterns for each constraint type|
|
Analyzes skill effectiveness data to identify failure patterns and recommend improvements. Use after /skill-monitor flags underperforming skills.
Run ad-hoc PostgreSQL analytics queries against dev/test database
Find and report technical debt in the codebase
|
|
Guide plugin development workflow — editing skills, agents, hooks, or eval framework in this repo. Use when modifying files in plugins/elixir-phoenix/, lab/eval/, or lab/autoresearch/. Ensures changes pass eval, lint, and tests before committing.