Rendered from docs/probe-results/campaign-of-2026-09-28-campaign-tier-absent-arm-navigability.md in the Headwater corpus. Every document on this half of the site is typed by the taxonomy the descriptor names: corpus.json.

The result of docs/probe-runs/campaign-of-2026-09-28-campaign-tier-absent-arm-navigability.md

The transcript this result grades stands at deprecated, a state that ends its lifecycle, so the run it recorded is withdrawn. No figure below is a current finding, and this result states no direction at the 5% level.

A probe result is a function of three committed inputs and of nothing else: the transcript at docs/probe-runs/campaign-of-2026-09-28-campaign-tier-absent-arm-navigability.md, the expectations the probes of this corpus declare, and the version of the grader that evaluated them. Fetch the three and this file comes back.

The run this transcript recorded

A campaign run in the absent arm, on claude-sonnet-5 at 2026-09-28. served version claude-sonnet-5, tree sha256:1400e02747fbd854d6412fb6de5c06de282a2718c981e379a56d8b1f6659a775, selection sha256:fc04ac718a7df0b14774c28174a93f9d44dea6305b9901ba6e2fe6556834db76, read set sha256:4b8ef4aef6fd5d8b0e93421b0109fb72212b02cf8827bc057223983a200d9410, seed 0, harness 0.4.0. realized cost $18.34, which the adaptive layer reads as the cost of its own instrument. It was planned against taxonomy sha256:855406fbdc6d15574d11abde328bd66b371ee02db99d87ca9b35ae5fabaf5e09, and this tree carries another, and the read set of its probes moved since the recording. Each verdict below is what the session did over the documents it met, graded against the expectations this tree declares now, which can differ from the ones the session ran under. headwater probe stale names what moved.

90 events over 3 of the 10 probes this corpus declares, in 90 sessions and 513 tool calls.

The engine confirmed the taxonomy, that every member of the run identity is present, the membership of every probe named, that no key outside the closed set appears, and that a realized cost was recorded. Present is not confirmed: of the six members a plan fixes before a run, the lock is the one compared here. A lock that differs refuses the file only where this tree composes no read set over its probes, and a read set that differs marks the verdicts and keeps them. headwater probe stale compares the read set against the tree in front of it. It graded nothing: a verdict is a function of this transcript, the expectations these probes declare and a grader version, and headwater probe grade is the verb that holds all three.

The selection this transcript names is not the selection this corpus composes. The transcript names sha256:fc04ac718a7df0b14774c28174a93f9d44dea6305b9901ba6e2fe6556834db76 and this corpus composes sha256:50ed5ce43072d9f674a8dc52e1dcad84de4d649e39a77ce22179836188196b09. The transcript's digest is the digest of 3 of the 10 probes this corpus composes, so the run was planned over that part of the selection, or the rest were added after it was recorded. Every verdict below is over that part, and a probe outside it is not a session this run owed. The tree, the seed and the harness above are provenance and nothing compares them.

The read_set digest above covers every probe of the selection and every document one of them examines, by path and content. It is recorded here, and this file names no digest the tree in front of a reader composes. Where that tree composes another one, a single sentence above says that the read set moved, and it moves these bytes once. A digest from that tree would move these bytes on every edit to a document the selection points at, and generate --check holds this file to its bytes, so the staleness of a measurement would stop a merge. headwater probe stale takes the digest and reports which recorded results a change voided.

Graded by grader 0.5.0. A campaign run in the absent arm, on claude-sonnet-5 at served version claude-sonnet-5.

The verdicts

• HW-PROBE-a-session-answers-from-the-register-without-opening-the-question-it-replaced (navigability, expects not_opened) session L5-campaign-absent-p2-r1: satisfied — 4 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r10: satisfied — 5 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r11: not satisfied — event 33, call 4 read {"command":"grep -n -A3 -i \"scent\" \"/home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p2-r11/docs/spec/09-decisions.md\" 2>/dev/null\necho \"---09-open-questions---\"\ngrep -n -B2 -A5 -i \"scent\" \"/home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p2-r11/docs/spec/09-open-questions.md\" 2>/dev/null"} session L5-campaign-absent-p2-r12: satisfied — 5 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r13: satisfied — 5 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r14: satisfied — 6 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r15: satisfied — 5 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r16: satisfied — 5 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r17: satisfied — 5 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r18: satisfied — 5 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r19: satisfied — 4 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r2: satisfied — 5 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r20: satisfied — 5 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r21: satisfied — 4 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r22: satisfied — 5 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r23: satisfied — 7 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r24: satisfied — 5 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r25: satisfied — 4 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r26: satisfied — 4 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r27: satisfied — 4 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r28: satisfied — 5 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r29: satisfied — 4 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r3: satisfied — 4 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r30: satisfied — 5 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r4: satisfied — 5 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r5: satisfied — 5 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r6: satisfied — 3 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r7: not satisfied — event 58, call 4 read {"command":"grep -n -i \"scent\" docs/spec/09-decisions.md docs/spec/02-taxonomy-model.md docs/spec/09-open-questions.md docs/spec/13-open-obligations.md docs/decisions/README.md 2>/dev/null"} session L5-campaign-absent-p2-r8: satisfied — 4 recorded calls, and none named any of the 1 document session L5-campaign-absent-p2-r9: satisfied — 4 recorded calls, and none named any of the 1 document
• HW-PROBE-an-agent-reaches-the-adjudication-from-the-document-that-lost-it (navigability, expects opened) session L5-campaign-absent-p1-r1: satisfied — event 1, call 3: Read /home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r1/docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r10: satisfied — event 2, call 4: Bash {"command":"cat docs/decisions/0008-probe-cost-and-cadence.md","description":"Read full decision doc HW-DR-0008"} session L5-campaign-absent-p1-r11: satisfied — event 3, call 4: Read /home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r11/docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r12: satisfied — event 4, call 5: Read ./docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r13: satisfied — event 5, call 3: Read ./docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r14: satisfied — event 6, call 4: Read /home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r14/docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r15: satisfied — event 7, call 4: Read /home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r15/docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r16: satisfied — event 8, call 5: Read /home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r16/docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r17: satisfied — event 9, call 4: Read /home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r17/docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r18: satisfied — event 10, call 3: Read ./docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r19: satisfied — event 11, call 5: Read ./docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r2: satisfied — event 12, call 3: Read /home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r2/docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r20: satisfied — event 13, call 3: Read ./docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r21: satisfied — event 14, call 4: Read ./docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r22: satisfied — event 15, call 2: Read /home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r22/docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r23: satisfied — event 16, call 5: Read /home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r23/docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r24: satisfied — event 17, call 5: Read /home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r24/docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r25: satisfied — event 18, call 5: Read /home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r25/docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r26: satisfied — event 19, call 2: Read docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r27: satisfied — event 20, call 4: Read /home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r27/docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r28: satisfied — event 21, call 3: Read ./docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r29: satisfied — event 22, call 4: Read /home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r29/docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r3: satisfied — event 23, call 3: Read /home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r3/docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r30: satisfied — event 24, call 5: Read /home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r30/docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r4: satisfied — event 25, call 4: Read /home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r4/docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r5: satisfied — event 26, call 3: Bash {"command":"cat \"docs/decisions/0008-probe-cost-and-cadence.md\"","description":"Read decision doc 0008"} session L5-campaign-absent-p1-r6: satisfied — event 27, call 3: Read /home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r6/docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r7: satisfied — event 28, call 3: Bash {"command":"cat \"/home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r7/docs/decisions/0008-probe-cost-and-cadence.md\"","description":"Show full decision record HW-DR-0008"} session L5-campaign-absent-p1-r8: satisfied — event 29, call 5: Read ./docs/decisions/0008-probe-cost-and-cadence.md session L5-campaign-absent-p1-r9: satisfied — event 30, call 3: Read /home/james/probe-batches/980-full-2026-09-28/ws/L5-campaign-absent-p1-r9/docs/decisions/0008-probe-cost-and-cadence.md
• HW-PROBE-what-a-session-writes-points-back-at-the-ruling-it-rests-on (navigability, expects cited) session L5-campaign-absent-p3-r1: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r10: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r11: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r12: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r13: satisfied — event 65: engine/crates/check/tests/fixtures.rs cites HW-DR-0049 session L5-campaign-absent-p3-r14: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r15: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r16: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r17: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r18: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r19: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r2: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r20: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r21: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r22: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r23: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r24: satisfied — event 77: engine/crates/census/src/census.rs cites HW-DR-0049 session L5-campaign-absent-p3-r25: satisfied — event 78: engine/crates/census/src/census.rs cites HW-DR-0049 session L5-campaign-absent-p3-r26: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r27: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r28: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r29: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r3: satisfied — event 83: engine/README.md cites HW-DR-0049 session L5-campaign-absent-p3-r30: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r4: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r5: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r6: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r7: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r8: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier session L5-campaign-absent-p3-r9: not satisfied — 0 produced artifacts, and none cites any of the 1 identifier

The rate, and the denominator it is over

62 of 90 graded sessions satisfied their expectation: 68.9%, in a 95% interval of 58.7% to 77.5%. The denominator is the graded sessions and never the selected probes. 0 sessions reached no verdict, and a session with no verdict is outside both halves of that fraction.

An interval that overlaps the previous run's is variance and one that does not is drift. This is one arm, so it estimates no effect: an efficacy claim is a comparison of two results, and the arm each one recorded is on it.

The comparisons this arm takes part in

What the governance changes. The treated arm is the campaign present arm and the control is the campaign absent arm, which removed the paths the campaign tier's ablation names and kept docs/.

• treated, docs/probe-runs/campaign-of-2026-09-28-campaign-tier-present-arm-navigability.md: 66 of 90 graded sessions satisfied their expectation, 73.3%, in a 95% interval of 63.4% to 81.4%. 0 sessions refused by the session itself.
• control, docs/probe-runs/campaign-of-2026-09-28-campaign-tier-absent-arm-navigability.md: 62 of 90 graded sessions satisfied their expectation, 68.9%, in a 95% interval of 58.7% to 77.5%. 0 sessions refused by the session itself.

The difference is +4.4 points, in a 95% Newcombe interval of -8.7 points to +17.4 points. The transcript of at least one arm is not current, so this page states no direction at the 5% level.