Shared memory and context tools for agentic work.
Code Rooms
{
"schema": "m1nd-bug-hunt-report-v0",
"round_id": "bughunt-p-limit-short-audit-20260515T113006Z",
"generated_at": "2026-05-15T11:35:21.321905+00:00",
"repo": "p-limit",
"source_commit": "9f52583119f0cb0d85c6fec600c94a21fd89d060",
"seeded_bug_count": 5,
"seeded_bug_ids": [
"options-object-default-rejects-on-clear",
"reject-on-clear-falsy-non-boolean-accepted",
"map-non-array-iterable-index-lost",
"limit-function-drops-arguments",
"infinite-concurrency-rejected"
],
"lanes_completed": 4,
"lanes_expected": 4,
"lanes": [
"lane_id": "audit-01",
"instruction_mode": "m1nd-short-audit",
"completed": true,
"result_schema": "m1nd-bug-hunt-audit-result-v0",
"result_path": "/Users/kle1nz/m1nd/docs/benchmarks/bug-hunt-rounds/bughunt-p-limit-short-audit-20260515T113006Z/lane-results/audit-01.json",
"events_path": "/Users/kle1nz/m1nd/docs/benchmarks/bug-hunt-rounds/bughunt-p-limit-short-audit-20260515T113006Z/event-streams/audit-01.jsonl",
"findings_count": 4,
"matched_seeded_bug_ids": [
"infinite-concurrency-rejected",
"options-object-default-rejects-on-clear"
"matched_findings": [
"finding_index": 0,
"finding_title": "limitFunction drops every argument passed to the returned wrapper",
"seeded_bug_id": "limit-function-drops-arguments"
},
"finding_index": 1,
"finding_title": "Options object changes rejectOnClear default and accepts falsey non-booleans",
"seeded_bug_id": "options-object-default-rejects-on-clear"
"finding_index": 2,
"finding_title": "limit.map passes undefined index for non-array iterables",
"seeded_bug_id": "map-non-array-iterable-index-lost"
"finding_index": 3,
"finding_title": "validateConcurrency rejects previously supported Infinity concurrency",
"seeded_bug_id": "infinite-concurrency-rejected"
}
"seeded_recall_count": 4,
"seeded_recall_rate": 0.8,
"missed_seeded_bug_ids": [
"reject-on-clear-falsy-non-boolean-accepted"
"extra_unadjudicated_findings_count": 0,
"event_count": 4,
"agent_event_count": 3,
"first_event_at": "2026-05-15T11:30:38.286062+00:00",
"first_agent_event_at": "2026-05-15T11:31:53+00:00",
"last_agent_event_at": "2026-05-15T11:34:37+00:00",
"agent_wall_clock_seconds": 164.0,
"assignment_to_first_agent_event_seconds": 74.714,
"first_finding_event_elapsed_seconds": 85.0,
"first_seeded_finding_event_elapsed_seconds": 85.0,
"timestamped_event_count": 4,
"timestamped_agent_event_count": 3,
"m1nd_usage_count": 2,
"agent_testimony": "I did not read operator-only artifacts and did not patch the audited repo. m1nd was used only for the bounded short-audit orientation pass. Final claims are based on direct source inspection, local git diff, and focused Node runtime probes."
"lane_id": "audit-02",
"result_path": "/Users/kle1nz/m1nd/docs/benchmarks/bug-hunt-rounds/bughunt-p-limit-short-audit-20260515T113006Z/lane-results/audit-02.json",
"events_path": "/Users/kle1nz/m1nd/docs/benchmarks/bug-hunt-rounds/bughunt-p-limit-short-audit-20260515T113006Z/event-streams/audit-02.jsonl",
"findings_count": 5,
"finding_title": "limitFunction drops all caller arguments",
"finding_title": "limit.map omits indexes for non-array iterables",
"finding_title": "Options-object form enables rejectOnClear by default despite documented default false",
"finding_title": "rejectOnClear accepts falsey non-boolean values",
"seeded_bug_id": "reject-on-clear-falsy-non-boolean-accepted"
"finding_index": 4,
"finding_title": "Infinity is rejected as a concurrency limit",
"seeded_recall_count": 5,
"seeded_recall_rate": 1.0,
"missed_seeded_bug_ids": [],
"event_count": 5,
"agent_event_count": 4,
"first_event_at": "2026-05-15T11:30:38.286333+00:00",
"first_agent_event_at": "2026-05-15T11:31:46+00:00",
"last_agent_event_at": "2026-05-15T11:38:25+00:00",
"agent_wall_clock_seconds": 399.0,
"assignment_to_first_agent_event_seconds": 67.714,
"first_finding_event_elapsed_seconds": 186.0,
"first_seeded_finding_event_elapsed_seconds": 186.0,
"timestamped_event_count": 5,
"timestamped_agent_event_count": 4,
"m1nd_usage_count": 1,
"agent_testimony": "I did not read operator-only artifacts and did not patch the audited repo. I wrote only the lane result JSON and appended lane event JSONL records. The local `npm test` command could not run because `xo` was not available on PATH (`sh: xo: command not found`), so behavioral verification used focused `node --input-type=module` probes."
"lane_id": "audit-03",
"instruction_mode": "direct",
"result_path": "/Users/kle1nz/m1nd/docs/benchmarks/bug-hunt-rounds/bughunt-p-limit-short-audit-20260515T113006Z/lane-results/audit-03.json",
"events_path": "/Users/kle1nz/m1nd/docs/benchmarks/bug-hunt-rounds/bughunt-p-limit-short-audit-20260515T113006Z/event-streams/audit-03.jsonl",
"finding_title": "Options-object construction silently enables rejectOnClear by default",
"finding_title": "limitFunction drops every argument passed to the limited wrapper",
"finding_title": "limit.map passes undefined as the index for non-array iterables",
"finding_title": "Falsy non-boolean rejectOnClear values bypass runtime validation",
"first_event_at": "2026-05-15T11:30:38.286565+00:00",
"first_agent_event_at": "2026-05-15T11:31:41+00:00",
"last_agent_event_at": "2026-05-15T11:33:57+00:00",
"agent_wall_clock_seconds": 136.0,
"assignment_to_first_agent_event_seconds": 62.713,
"first_finding_event_elapsed_seconds": 61.0,
"first_seeded_finding_event_elapsed_seconds": 61.0,
"m1nd_usage_count": 0,
"agent_testimony": "Direct Mode honored: I did not use m1nd tools, m1nd helper scripts, or operator-only artifacts, and I did not patch the audited repo. I wrote only the required lane result JSON and appended required agent events to the lane event JSONL."
"lane_id": "audit-04",
"result_path": "/Users/kle1nz/m1nd/docs/benchmarks/bug-hunt-rounds/bughunt-p-limit-short-audit-20260515T113006Z/lane-results/audit-04.json",
"events_path": "/Users/kle1nz/m1nd/docs/benchmarks/bug-hunt-rounds/bughunt-p-limit-short-audit-20260515T113006Z/event-streams/audit-04.jsonl",
"finding_title": "Options object enables rejectOnClear by default despite the documented false default",
"finding_title": "Falsey non-boolean rejectOnClear values bypass runtime validation",
"finding_title": "limitFunction drops all caller arguments before invoking the wrapped function",
"finding_title": "validateConcurrency rejects Infinity although the previous implementation allowed it",
"first_event_at": "2026-05-15T11:30:38.286782+00:00",
"last_agent_event_at": "2026-05-15T11:33:20+00:00",
"agent_wall_clock_seconds": 99.0,
"first_finding_event_elapsed_seconds": 69.0,
"first_seeded_finding_event_elapsed_seconds": 69.0,
"agent_testimony": "Direct Mode followed: no m1nd tools and no m1nd helper scripts were used. I did not patch the audited repo. npm test could not run because the workspace lacked node_modules/.bin/xo, so I used focused Node runtime probes against the checked-out index.js plus source/documentation/type inspection."
"arms": {
"direct": {
"lane_count": 2,
"completed_lane_count": 2,
"seeded_bug_count_per_lane": 5,
"seeded_recall_total": 9,
"seeded_possible_total": 10,
"seeded_recall_rate": 0.9,
"per_lane_seeded_recall_counts": [
4,
5
"median_seeded_recall_count": 4.5,
"average_seeded_recall_count": 4.5,
"median_agent_wall_clock_seconds": 117.5,
"median_first_finding_event_elapsed_seconds": 65.0,
"median_first_seeded_finding_event_elapsed_seconds": 65.0,
"total_findings": 9,
"extra_unadjudicated_findings_total": 0,
"audit-03",
"audit-04"
]
"m1nd-short-audit": {
"median_agent_wall_clock_seconds": 281.5,
"median_first_finding_event_elapsed_seconds": 135.5,
"median_first_seeded_finding_event_elapsed_seconds": 135.5,
"audit-01",
"audit-02"
"comparability": {
"all_lane_results_present": true,
"primary_arm_lane_counts": {
"direct": 2,
"m1nd-short-audit": 2
"rate_comparison_available": true,
"balanced_lane_counts": true,
"comparability_notes": [
"Compare rates rather than raw totals when arm lane counts differ.",
"Extra findings are unadjudicated and are not used as precision penalties."
"top_line": {
"median_seeded_recall_count": 4.5
"invalidated_attempts": [],
"public_claim_worthy": false,
"public_claim_blockers": [
"single internal round",
"one fixture repo",
"seeded recall only; extra findings are not independently judged"
"non_claims": [
"Finding extra real issues is allowed but seeded recall is measured against these five defects.",
"Primary auditors are not told bug count or comparison arm.",
"This round tests p-limit seeded behavioral defects, not universal agent performance.",
"agent testimony is not evidence without scored finding artifacts",
"extra findings are reported as unadjudicated, not as false positives",
"m1nd does not replace tests, compiler output, git history, rg, or direct file truth",
"one bug-hunt round is not a public performance claim",
"seeded recall does not measure all real defects in the fixture repo"