{"repo":"NousResearch/hermes-agent","public_id":"43547","display_label":"Account for output-token reservation in context compaction thresholds","summary":"Most members concern the compaction/compression trigger not factoring in reserved output tokens, with one issue noting a custom-provider default max_tokens of 65536 reducing usable input budget. One member differs slightly, addressing output-cap retry convergence to a workable max_tokens, which is related but a distinct change.","subsystem":"agent","counts":{"members":5,"open_issues":0,"open_issues_distinct":0,"open_prs":1,"closed_issues":1,"merged_prs":1,"attempt_prs":3},"landed_fix":{"status":"merged_to_main","basis":"structural","pr":50781,"merged_at":"2026-06-22T12:03:09+00:00"},"resolution":{"state":"resolved","by":"pr-50781"},"freshness":{"state":"verified","last_verified_at":"2026-06-22T12:03:09+00:00","is_stale":false},"grouping":{"backbone":"STRUCTURAL_SPANNING","link_density":0.4},"fix_edges":[{"issue_number":43547,"pr_number":50781,"tier":"verified","basis":"closing_ref","confidence":0.95,"recoverable_from_timeline":true,"pr_merged_at":"2026-06-22T12:03:09+00:00"}],"members":[{"kind":"issue","number":43547,"title":"Context compaction trigger ignores the output-token reservation (custom-provider default max_tokens=65536 halves the usable input budget)","url":null,"state":"closed"},{"kind":"pr","number":43651,"title":"fix(compress): account for output-token reservation in compaction threshold (#43547)","url":null,"state":"closed","close_reason":{"refs":[{"kind":"commit","value":"623b21bf2","number":0},{"kind":"pr","value":"#50781","number":50781},{"kind":"pr","value":"#43547","number":43547},{"kind":"pr","value":"#14690","number":14690}],"category":"merged_via","implication":"superseded"},"merged":false},{"kind":"pr","number":45351,"title":"fix(agent): reserve output tokens in compression threshold","url":null,"state":"closed","close_reason":{"refs":[{"kind":"pr","value":"#43651","number":43651}],"category":"superseded_by_pr","implication":"superseded"},"merged":false},{"kind":"pr","number":50781,"title":"fix(compress): reserve output tokens in the compaction threshold (#23767, #43547)","url":null,"state":"merged","close_reason":{"category":"merged"},"merged":true,"is_fix_pr":true,"fix_tier":"verified","files_changed":["agent/agent_init.py","agent/context_compressor.py","tests/agent/test_context_compressor.py"]},{"kind":"pr","number":89777,"title":"fix(agent): converge output-cap retry to a workable max_tokens on first attempt","url":null,"state":"open","close_reason":{"category":"uncategorized","implication":"needs_llm"},"merged":false}],"trust":{"default_tier":"verified","includes_suggested":false}}