{"repo":"NousResearch/hermes-agent","public_id":"62605","display_label":"Token estimate underestimation in auto-compression preflight checks","summary":"The members concern auto-compression where rough token estimates may underestimate actual prompt size, with one issue reporting the total exceeding context_length after 3 retries. Two pull requests propose adjustments (calibrating the preflight estimate against real usage, and adding a 15% margin to the system prompt estimate); their outcomes are not stated in the titles.","subsystem":"agent","counts":{"members":3,"open_issues":1,"open_issues_distinct":1,"open_prs":1,"closed_issues":0,"merged_prs":0,"attempt_prs":2},"landed_fix":null,"resolution":{"state":"unresolved","by":null},"freshness":{"state":"never_verified","last_verified_at":null,"is_stale":false},"grouping":{"backbone":"STRUCTURAL_SPANNING","link_density":0.6666666666666666},"fix_edges":[],"members":[{"kind":"issue","number":62605,"title":"Auto-compression fails when rough token estimate underestimates actual prompt — total exceeds context_length after 3 retries","url":null,"state":"open"},{"kind":"pr","number":62612,"title":"fix(compressor): add 15% safety margin to system prompt token estimate","url":null,"state":"open","close_reason":{"category":"uncategorized","implication":"needs_llm"},"merged":false},{"kind":"pr","number":62649,"title":"fix(compression): calibrate rough preflight estimate against real usage (#62605)","url":null,"state":"closed","close_reason":{"category":"uncategorized","implication":"needs_llm"},"merged":false}],"trust":{"default_tier":"verified","includes_suggested":false}}