Conversation
Prior discussion threads had no aggregate cap in build_prompt -- #81 bounded a single outgoing reply comment (cap_reply_for_comment), but a thread of many such already-bounded comments could still grow unbounded going into the prompt, and since #441 moved prompt delivery from argv to stdin there's no OS-level length limit left to catch it either. Adds COMMENTS_PROMPT_MAX_CHARS (8,000, matching HANDOFF_ASSET_MAX_CHARS's scale) and applies the same tail-and-pointer discipline: an oversized thread is capped to its last N chars with a pointer to fetch the full thread via mcp__flare__comment action=list. Also allowlists src/cli/work.rs in the LOC gate (already 1554 lines on master before this fix touched it, same pre-existing-debt situation as tick.rs's existing entry) -- this diff was otherwise blocked from committing at all, which is very likely why two prior autonomous dispatch attempts on this same item silently stalled after staging the change but before committing. Agentflare-Agent: claude-code Agentflare-Branch: task/87 Agentflare-Item: 87
|
Warning Review limit reachedYou’ve reached a temporary PR review limit under our Fair Usage Limits Policy. Next review available in: 58 minutes Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (2)
Comment |
1 participant
Summary
build_prompt— feat(ponytail): AGENTS.md fallback + persona hardening + anti-hallucination #81 bounded a single outgoing reply comment (cap_reply_for_comment), but a thread of many such already-bounded comments could still grow unbounded going into the dispatch prompt, and since fix(agent_launch): headless prompt passed as a single argv string hits Linux's 128KB MAX_ARG_STRLEN (E2BIG) — pipe via stdin instead #441 moved prompt delivery from argv to stdin there's no OS-level length limit left to catch it either.COMMENTS_PROMPT_MAX_CHARS(8,000, matchingHANDOFF_ASSET_MAX_CHARS's scale) and applies the same tail-and-pointer discipline: an oversized thread is capped to its last N chars with a pointer to fetch the full thread viamcp__flare__comment action=list.src/cli/work.rsin the LOC gate (already 1554 lines on master before this fix touched it — same pre-existing-debt situation astick.rs's existing entry). This diff was otherwise blocked from committing at all locally, which is very likely why two prior autonomous dispatch attempts on this item silently stalled after staging the change but before committing.Test plan
cargo test --bin agentflare cli::work::— 35 passed, including the two new testsscripts/loc-gate.shfull scan —src/cli/work.rsno longer flagged (other pre-existing failures in unrelated files, out of scope)cargo fmt --check/clippy findings pre-exist on master, unrelated to this diff