a11y fix agent stuck on duplicate id rewrite loop

a11y fix agent stuck on duplicate id rewrite loop - how to fix it

TL;DR

Rewrite ids deterministically in one pass: collect all duplicated ids, assign new unique values in a single edit, and verify. Agents that fix one duplicate, re-scan, and find the next one loop because rendering shifts ids between runs. One line of why: id duplication is a global property of the document, and fixing it incrementally re-rolls the dice each pass.

The error, verbatim

RemediationAgentError: duplicate id rewrite loop, 12 iterations
    fixed: submit-btn, then email-input appeared duplicated
    cause: component renders ids per mount, each fix shifts the set

Fix it step by step

Step 1: Reproduce the loop

node agent/remediate.js --rule duplicate-id-active | rg -i 'loop|iteration' | head -5

Expected: The agent fixes one id per iteration and new duplicates keep appearing.

Step 2: Collect the full duplicate set

npx @axe-core/cli https://example.com/ --rules duplicate-id-active --save axe-d.json | tail -2

Expected: One scan lists every duplicated id at once.

Step 3: Rewrite in a single pass

rg -n 'duplicate' agent/patches/duplicate-id.js | head -10

Expected: Change the patch to collect all duplicates from one scan and rewrite them together.

Step 4: Re-run remediation

node agent/remediate.js --rule duplicate-id-active | tail -4

Expected: Single pass clears all duplicates, verification passes.

Step 5: Add a regression probe

node agent/run-audit.js --smoke | tail -3

Expected: Smoke run passes; schedule it so the breakdown is caught if it ever regresses.

When to use this skill

• You run an agent that scans UIs for accessibility and it hits this breakdown
• The agent's scan loop stalls, crashes, or loops on this exact failure
• You are hardening an audit agent's error handling for production scans

When NOT to use this skill

• A human runs the scan manually and it works, this is agent-harness failure handling
• The scan completes and only reports violations, use the rule-specific skills

Compatibility

Remediation agent with source edit primitives. One axe scan provides the complete duplicate set. Pin the tool version in the lockfile so scans stay reproducible across machines.

Variant phrasings

agent duplicate id loop

Same loop, same single-pass fix.

fix agent keeps finding new duplicate ids

Practitioner phrasing.

the breakdown hits other routes too

Agent failure modes are systemic; apply the hardening to every route the agent covers, not just the one that failed.

Why it happens

Components that generate ids per mount (or hardcode them) produce a document-wide duplicate set that shifts as the agent edits: fixing submit-btn changes rendering enough that email-input now collides. Per-violation patch loops chase a moving target. The correct patch collects the entire duplicate set from one scan, assigns deterministic unique ids (suffixing with a stable component path), applies all edits, then verifies once. If duplicates persist after a single pass, the id generation itself is broken and needs a useId-style fix, not more rewriting. Agent breakdowns are systemic: the same failure mode will hit every route, page, or run the agent touches. Harden the harness once (timeouts, loop detection, verification gates) instead of patching per page, and keep breakdown telemetry separate from violation counts.

Edge cases

• Prefer fixing the id generation (unique id hooks) over rewriting rendered ids, rewriting is a bandage.
• Deterministic suffixes beat random ones, random ids break snapshot tests and reintroduce collisions.
• Verify with a fresh full-page scan, not the agent's incremental state.
• Log breakdowns separately from violations in agent telemetry; mixing them hides whether the agent itself is getting more reliable.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_ZoVs5utN3mupUDfs0XTKew