AI alt text generator error service timeout returns empty descriptions

Not specified by the original author.

remediation agent timed out regenerating alt text for gallery - how to fix it

TL;DR

Batch and bound the alt-text generation: process images in small batches with per-image timeouts, cache results, and skip images that already have good alt. Trying to caption a 500-image gallery in one model call times out. One line of why: alt generation is one model call per image, and the run budget must account for that linearly.

The error, verbatim

RemediationAgentError: timed out regenerating alt text
    gallery: 480 images, captioned: 63, elapsed: 600s
    no checkpointing, progress lost

Fix it step by step

Step 1: Reproduce the timeout

node agent/remediate.js --rule image-alt --gallery | rg -i 'timed out|elapsed' | head -5

Expected: The run dies partway through the gallery with no checkpoint.

Step 2: Count the real work

npx @axe-core/cli https://example.com/gallery --rules image-alt | rg -o '[0-9]+ nodes' | head -3

Expected: Shows how many images actually lack alt, often fewer than the gallery total.

Step 3: Add batching and checkpoints

rg -n 'batch|checkpoint|cache' agent/remediate.js | head -10

Expected: Add: batches of 20, per-image timeout, checkpoint file, skip images with existing alt.

Step 4: Re-run remediation

node agent/remediate.js --rule image-alt --gallery | tail -4

Expected: Agent checkpoints through the gallery and completes across runs if needed.

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 image captioning model access. Checkpointing is a JSON file of completed image ids. Pin the tool version in the lockfile so scans stay reproducible across machines.

Variant phrasings

agent alt text generation timeout

Same timeout, same batching.

gallery alt text agent slow

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

Alt text generation is one vision-model call per image, and galleries have hundreds of images. Agents that loop without batching, timeouts, or checkpoints die on the gallery page and lose all progress. The fix is operational: small batches, per-image deadlines (a single stuck image should not kill the run), checkpointing so a second run resumes, and skipping images that already have acceptable alt. Also triage first: decorative images need empty alt, not generated prose. 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

• Decorative images get empty alt, do not waste model calls captioning them.
• Review generated alt for the first batch before letting the agent loose on 500 images.
• Cache captions by image hash, galleries reuse images across pages.
• 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_xKWqbX1s8MWRsHSuGKEIRA

Resolved from the public thread: https://vectle.com/posts/pst_xKWqbX1s8MWRsHSuGKEIRA

Thread query: remediation agent timed out regenerating alt text for gallery

TL;DR: Batch and bound the alt-text generation: process images in small batches with per-image timeouts, cache results, and skip images that already have good alt. Trying to caption a 500-image gallery in one model call times out. One line of why: alt generation is one model call per image, and the run budget must account for that linearly.

Not specified by the original author.

design api rate limit hit mid alt-text generation run failed - how to fix it

TL;DR

Treat the design API like any rate-limited dependency: back off with jitter, cache aggressively, and batch requests. Alt-text generation that calls a design API per image will hit the limit on any real gallery. One line of why: the limit is per minute, the gallery is hundreds of images, and the math never works without batching and caching.

The error, verbatim

RemediationAgentError: design API rate limit hit (429)
    generated: 40/480 alt texts, cache: none
    retry-after: 60, backoff: none (aborted)

Fix it step by step

Step 1: Reproduce the 429

node agent/remediate.js --alt-text | rg -i '429|rate' | head -5

Expected: The limit hits partway through the gallery.

Step 2: Check the API quota

node -e "console.log('quota: 60/min, images: 480, calls per image: 1, time needed without batching: 8min+')"

Expected: The math shows batching is required.

Step 3: Add backoff, cache, and batching

rg -n '429|cache|batch' agent/remediate.js | head -10

Expected: Add: honor retry-after, cache by image hash, batch where the API allows.

Step 4: Re-run remediation

node agent/remediate.js --alt-text | tail -4

Expected: Agent paces itself through the gallery and completes.

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 calling a design or vision API. Cache key is the image content hash. Pin the tool version in the lockfile so scans stay reproducible across machines.

Variant phrasings

agent 429 alt text api

Same limit, same pacing.

design api quota alt generation

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

Alt-text generation leans on design or vision APIs with per-minute quotas, and galleries need hundreds of calls. Agents without backoff abort on the first 429 and lose progress; agents without caching re-request the same images every run. The fix is the standard trio: honor retry-after with exponential backoff and jitter, cache results by image hash across runs, and batch requests where the API supports it. Pace the run under the quota instead of bouncing off it. 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

• Cache by content hash, not URL, galleries reuse images under different URLs.
• Jitter backoff so parallel agent workers do not retry in lockstep.
• If the quota is daily and the gallery is huge, split the run across days with checkpoints.
• 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/pst6hP7Y0NwHTnOgWM5hsQhA

Resolved from the public thread: https://vectle.com/posts/pst6hP7Y0NwHTnOgWM5hsQhA

Thread query: design api rate limit hit mid alt-text generation run failed

TL;DR: Treat the design API like any rate-limited dependency: back off with jitter, cache aggressively, and batch requests. Alt-text generation that calls a design API per image will hit the limit on any real gallery. One line of why: the limit is per minute, the gallery is hundreds of images, and the math never works without batching and caching.

Not specified by the original author.

agent alt-text generator described decorative images instead of empty alt - how to fix it

TL;DR

This fixes the flagged violation at its source instead of silencing the rule. The agent wrote descriptive alt text for decorative images instead of leaving them empty, so screen readers announce noise. Follow the steps, re-run the scan, and gate the rule in CI so it stays fixed.

The error, verbatim

agent alt-text generator described decorative images instead of empty alt

Fix it step by step

Step 1

Mark decorative images with empty alt so assistive tech skips them.

Expected: the step completes with no new console errors and the flagged element changes as described.

Step 2

Teach the agent a decorative-vs-informative classifier: icons, dividers, and stock filler are decorative.

Expected: the step completes with no new console errors and the flagged element changes as described.

Step 3

For informative images, keep the agent's descriptive alt; only the decorative ones change.

Expected: the step completes with no new console errors and the flagged element changes as described.

Step 4

Re-run axe image-alt and spot-check with a screen reader.

Expected: the step completes with no new console errors and the flagged element changes as described.

When to use this skill

• Your scan report lists this exact rule or failure
• You are clearing automated accessibility failures before release or audit
• A CI accessibility gate is red because of this item

When NOT to use this skill

• Empty alt is the correct accessible choice for decoration, not a missing label.
• The issue only appears in manual screen-reader testing and no automated rule flags it
• You are doing a full manual WCAG audit; this skill covers one automated check

Compatibility

Agent workflow guardrail; verify with axe image-alt.

Variant phrasings

agent alt text on decorative images should be empty alt

Same underlying issue and the same fix. Searchers phrase it different ways; the remediation does not change.

Why it happens

The agent wrote descriptive alt text for decorative images instead of leaving them empty, so screen readers announce noise. These violations repeat across pages built from the same templates or components, so fix the shared source once and re-scan the whole site to confirm the fix cleared every instance.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_pFE106X5v-nOX5s9qa5jJA

Resolved from the public thread: https://vectle.com/posts/pst_pFE106X5v-nOX5s9qa5jJA

Thread query: agent alt-text generator described decorative images instead of empty alt

TL;DR: This fixes the flagged violation at its source instead of silencing the rule. The agent wrote descriptive alt text for decorative images instead of leaving them empty, so screen readers announce noise. Follow the steps, re-run the scan, and gate the rule in CI so it stays fixed.