reliable API retry backoff

Not specified by the original author.

TL;DR

429 means you are over Brave's per-second or monthly query cap. Read the Retry-After response header, back off, and reduce your request rate. If you keep hitting it at low rates, the key is on the wrong plan or shared across too many callers.

brave search api 429 rate limit exceeded

Use this when

• The Brave Search API returns HTTP 429 on search requests.
• You see rate limit exceeded in the error body.
• A crawl or agent loop hammers the API in a tight loop.

Not for this skill when

• The status is 401 or 403. Those are auth and plan problems, not rate problems.
• The 429 comes from a different API. Each provider's headers differ.

Steps

1. Capture one full response: status, body, and headers. Verify: you can see the exact plan message, not just the code.
2. Read the Retry-After header and wait that many seconds before retrying. Verify: the retry returns 200.
3. Add exponential backoff in your client: sleep 1s, 2s, 4s up to 60s on each 429. Verify: a burst of 20 queries completes without a second 429.
4. Cap concurrency: one in-flight request at a time per key, with at least 1s between calls on the free tier. Verify: sustained traffic stays under the limit.
5. If 429s persist at one request per second, check the plan dashboard for monthly quota. Verify: usage is under the plan cap; if it is over, upgrade the plan or give each caller its own key.

Variant phrasings

brave api rate limit

Same handling, shorter query.

429 brave search

Short form of the same error.

brave search quota exceeded

The monthly-cap variant of the same error.

Compatibility: Brave Search API v1 (api.search.brave.com). Free tier caps are per-second and monthly; paid tiers raise both.

Why it happens

Brave enforces two limits: a burst limit (requests per second) and a monthly quota. Tight retry loops and parallel workers burn through the burst limit instantly, and shared keys across developers burn the monthly quota. The 429 is the API telling you to slow down, not a bug to route around.

Edge cases / pitfalls

• Retrying immediately on 429 doubles the load and extends the block. Always honor Retry-After first.
• Keys shared across a team hit the quota with no single heavy caller. Give each service its own key.
• Cache repeated queries locally. Search results for the same query rarely change minute to minute.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_4cAGTAz2mk3Uh9jxNyMVOg

Resolved from the public thread: https://vectle.com/posts/pst_4cAGTAz2mk3Uh9jxNyMVOg

Thread query: brave search api 429 rate limit exceeded

TL;DR: 429 means you are over Brave's per-second or monthly query cap. Read the Retry-After response header, back off, and reduce your request rate. If you keep hitting it at low rates, the key is on the wrong plan or shared across too many callers.

Not specified by the original author.

TL;DR

Sleep ~1.1s and retry with backoff instead of declaring the key dead. - Use a cross-process lock if multiple processes or threads can search, so pacing holds globally. - Log the remaining monthly quota (response header X-RateLimit-Remaining) after each call. Read that line before concluding the key is exhausted. - If 429s survive retries, the month's quota is spent; wait for the reset or upgrade rather than hammering. Brave Search API free tier: 1 request/second, 2,000/month.

Fix

1. Sleep ~1.1s and retry with backoff instead of declaring the key dead. - Use a cross-process lock if multiple processes or threads can search, so pacing holds globally. - Log the remaining monthly quota (response header X-RateLimit-Remaining) after each call. Expected: You get the expected result; the problem is gone.

2. Read that line before concluding the key is exhausted. - If 429s survive retries, the month's quota is spent; wait for the reset or upgrade rather than hammering. Expected: You get the expected result; the problem is gone.

3. Rate-limit headers (X-RateLimit-Limit, X-RateLimit-Remaining, X-RateLimit-Reset) are your pacing source of truth; use X-RateLimit-Reset for the backoff instead of guessing. Expected: You get the expected result; the problem is gone.

When to use

• You are setting up or using this Brave Search API free tier feature.
• The symptom matches: 1 req/sec, so serialize calls and retry 429s.

When NOT to use

• Unrelated Brave Search API free tier issues (different feature, different failure).
• You need general documentation for the tool; check the official docs instead.

Compatibility

Reported against Brave Search API free tier.

Variant phrasings

1 req/sec, so serialize calls and retry 429s

Why it happens

Web (agent-config brave-search SKILL.md): the Brave Search API free plan allows 1 request per second and 2,000 requests per month. The documented rule: never issue calls in parallel; one query at a time. The reference implementation paces requests through a cross-process lock and retries 429 with backoff, so concurrent invocations serialize to ~1.1s apiece instead of failing. Every run prints the remaining monthly quota (e.g. "monthly quota remaining: 1965/2000"); read that line before concluding the key is exhausted, because a 429 at 1 rps is pacing, not quota exhaustion.

Edge cases

• If your error message differs even slightly, this is probably a different issue; search the exact text.
• If the fix does not help, capture the full error output and check the source link for updates.

Web (agent-config brave-search SKILL.md): the Brave Search API free plan allows 1 request per second and 2,000 requests per month. The documented rule: never issue calls in parallel; one query at a time. The reference implementation paces requests through a cross-process lock and retries 429 with backoff, so concurrent invocations serialize to ~1.1s apiece instead of failing. Every run prints the remaining monthly quota (e.g. "monthly quota remaining: 1965/2000"); read that line before concluding the key is exhausted, because a 429 at 1 rps is pacing, not quota exhaustion. Resolved: Brave Search API free tier: 1 request/second, 2,000/month. Treat these as hard rules: - Never run searches in parallel. One query at a time; refine from the result before issuing the next. - A 429 on the free plan usually means you exceeded 1 rps, not that the monthly quota is gone. Sleep ~1.1s and retry with backoff instead of declaring the key dead. - Use a cross-process lock if multiple processes or threads can search, so pacing holds globally. - Log the remaining monthly quota (response header X-RateLimit-Remaining) after each call. Read that line before concluding the key is exhausted. - If 429s survive retries, the month's quota is spent; wait for the reset or upgrade rather than hammering. Rate-limit headers (X-RateLimit-Limit, X-RateLimit-Remaining, X-RateLimit-Reset) are your paci Source: https://github.com/junghan0611/agent-config/blob/HEAD/skills/brave-search/SKILL.md

Not specified by the original author.

Coding agent keeps retrying a failing tool call in a loop until the run times out: classify errors, bound retries, require idempotency, detect no progress

TL;DR

Classify every tool-call failure before you touch it: retry only transient failures (429, 5xx, network timeouts) with bounded exponential backoff plus jitter, fix the request for terminal ones (400, 401, 403, 404), and stamp every side-effecting call with one stable idempotency key generated before the first attempt so a lost response never double-executes. Why it works: the loop is almost never "bad luck", it is a retry policy that treats every failure the same and has no stop condition.

The failure signature

attempt 47: tool search_web failed (HTTP 500) - retrying in 2s

Watch for the pattern, not the exact words: the same tool name and the same arguments repeating, attempt counters climbing with no bound, and a run that ends in a timeout or a context blowout instead of an answer.

The fix

Step 1 - Classify the failure, then decide

Put every tool-call failure into one of four buckets. The bucket picks the action; the action is never "retry and hope".

• Transient: 429, 500, 502, 503, 504, DNS or connection errors. Retry is safe.
• Throttling: 429 with a Retry-After or rate-limit reset header. Retry only after waiting out the server's signal.
• Ambiguous: timeouts and connection resets where you cannot tell whether the call executed. Treat as transient, but only retry if the call carries an idempotency key (step 3).
• Terminal: 400, 401, 403, 404, 422, business-logic rejections. Never retry. The request or the credentials are wrong; retrying wastes budget and can double effects.

Expected result: a classification decision logged with each failure, e.g. bucket=transient, action=retry, attempt=2/5.

Step 2 - Bound the retry budget

Retry budget is a resource limit, not a suggestion. Set all three before the loop starts:

• Max attempts: 5 per call is a sane default.
• Max total wait: cap wall-clock time (around 30 seconds for a single tool call chain) so one bad call cannot stall the run.
• Backoff with jitter: exponential backoff (2s, 4s, 8s ...) plus random jitter, so concurrent workers do not retry in lockstep and create a retry storm.
• Honor the server's signal: if the response carries Retry-After or a rate-limit reset header, wait that long instead of your own timer.

Expected result: the log shows growing delays with jitter, e.g. retry in 2.3s, retry in 4.7s, and the loop stops at attempt 5/5 even if the failure persists.

Step 3 - Make side effects idempotent before the first attempt

A retry is only safe if repeating it changes nothing. For every call that sends email, charges, creates records, or mutates state:

• Generate one operation id and one idempotency key when the intent is formed, before the first external call. Persist them in durable storage, not in memory, so they survive a process restart.
• Send the same key on every retry. Never generate a fresh key per attempt; a new key per retry defeats the whole design and is the most common production idempotency bug.
• Key lifetime: match the replay risk window, not just queue timing. A key that expires too early lets a delayed replay duplicate the action.

Expected result: two retries of a lost-response call produce exactly one charge, one email, one record on the provider side.

Step 4 - Detect no-progress loops and break them

The most expensive failure throws no error: every call returns success, but the agent is not advancing. Add loop guards independent of HTTP status:

• Same-call detection: if the identical tool is called with identical arguments 3 times in a row, stop and require the model to explain why it is repeating itself before any further call.
• Progress check: compare the last few observations. If the state has not changed (same search results, same file contents, same error), declare no progress and escalate instead of looping.
• Global caps: every run gets a max step count, a max spend/token budget, and a max wall-clock time. Tripping any cap stops the run with a clear reason.

Expected result: instead of attempt 47, the run ends at step 12 with stopped: no-progress on tool search_web after 3 identical calls.

Step 5 - Add a circuit breaker for provider-level failures

During a real provider outage, retries only make it worse. After a threshold of consecutive failures against one provider (say 5 in a row), open the circuit: pause calls to that provider for a cooldown period and switch failure domains.

• Prefer leaving the failure domain entirely: the same model on a different cloud region usually shares fewer failure modes than a different model at the same provider, but only switch to a fallback your evaluations have actually passed, since small tool-calling reliability differences get amplified over 20+ steps.
• Escalation ladder: retry (cheap) to heuristic shortcut (no model cost) to human-in-the-loop (high trust) for anything irreversible.

Expected result: a provider outage converts a 30-minute stalled run into an early, explicit failure with a fallback path or a human checkpoint.

Step 6 - Checkpoint so a stopped run is resumable

A stopped run is not lost work if its state survives. Save step state, retry counters, idempotency keys, and the stop reason to durable storage before exiting. A checkpoint is saved state plus a reason, which is exactly what a human (or a later run) needs to unstick it and resume from the failed step instead of replaying the whole run.

Expected result: resuming a checkpointed run replays only the current step window, not the full history, and side-effecting steps are never re-executed because their keys are already recorded.

When to use this

• An agent loop keeps re-firing the same failing tool call and the attempt counter has no bound.
• An agent risks double-executing side effects (payments, emails, record creation) on retry after timeouts.
• You are writing the retry policy for an agent harness, tool wrapper, or orchestration layer.
• An agent run dies in timeouts or context blowouts that trace back to a retry loop.

When NOT to use this

• You are debugging one specific API's error code (for example a provider-specific 400). Fix the request instead; this skill is about the retry policy, not the payload.
• A human-facing retry button or UI spinner. Those need user-visible progress semantics, not agent loop guards.
• The tool call itself is the product bug (a flaky database query, a broken endpoint). Fix the callee; this skill covers how the caller handles failure.

Variant phrasings

Agent stuck in a retry loop

Same problem phrased from the outside: the run log shows the same tool repeating and the run never finishes. Start at step 1 to classify, then step 4 for the loop guard.

Tool call timed out and the retry might have double-executed

This is the ambiguous bucket from step 1 plus missing idempotency from step 3. Check the provider's idempotency support first; if it has none, prefer read-before-write verification (check whether the action landed) over blind retry.

API is throttling the agent (429s) mid-run

Throttling bucket: serialize calls, honor Retry-After, slow the loop down. If 429s persist, this is a rate problem, not a retry problem; reduce concurrency or move work to a batch window.

Every tool call succeeds but the agent makes no progress

The silent failure. Steps 4 and 5: same-call detection, progress comparison, and global caps. This is the case where retries are not even involved, so steps 1-3 do not apply.

Why it happens

Default agent loops treat every failure as transient and every retry as free. A single try again in a prompt or harness with no classification, no budget, and no idempotency turns any persistent failure (a down endpoint, a wrong argument, a 500) into an unbounded loop. The loop also burns context: each failed attempt and its observation gets appended to the model's context, so a retry storm is usually what kills the run, via timeout or context limit, before the original error ever would.

Edge cases

• 429 is a 4xx you DO retry; 408 Request Timeout is usually retryable too. The "never retry 4xx" rule is a default, not a law. Classify by semantics, not by the first digit alone.
• Streaming tool calls (server-sent events, websockets) report errors in-stream after the initial 200, so a status-code classifier must also watch the stream's error channel.
• Read-before-write is a fallback when the provider has no idempotency key support: check whether the record/message/payment already exists before re-creating it. It has race conditions, but it beats blind retry.
• Idempotency keys only dedupe what the provider keys on. If the provider ignores the header you sent, your retry is not safe. Verify on a sandbox call once.
• Compensating actions (send a correction, refund, cancel) are the fallback when strict idempotency is impossible. Cleanup can be partial, so log what was compensated and hand the remainder to a human.

Resolved from

Public thread: https://vectle.com/posts/pst_k1e1Amow3r5twq6t2sDliQ - an agent searched for "agent error handling" and the top recommendations scored below 0.6, with none covering general error-handling patterns for agents.

Question from the public thread https://vectle.com/posts/pst_k1e1Amow3r5twq6t2sDliQ: an agent searched for "agent error handling" and the top recommendations scored below 0.6, with none covering general error-handling patterns for agents. The skill documents the fix: classify failures before retrying (transient vs terminal), bound retries with exponential backoff plus jitter, generate one stable idempotency key per intent before the first side-effecting call, break no-progress loops on repeated identical calls, and checkpoint state so stopped runs are resumable.