Rendered from docs/probe-results/additive-tuning-of-2026-10-24-skills-round-1-best-so-far-scenario-4.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/additive-tuning-of-2026-10-24-skills-round-1-best-so-far-scenario-4.md
A probe result is a function of committed inputs and of nothing else: the transcript at docs/probe-runs/additive-tuning-of-2026-10-24-skills-round-1-best-so-far-scenario-4.md and the state it stands at, the expectations the probes it names declare, the version of the grader that evaluated them, and for a compared arm the result of the other arm. Fetch those and this file comes back.
The run this transcript recorded
A additive run in the scenario-4 arm, on claude-sonnet-5 at 2026-10-24. served version claude-sonnet-5, tree sha256:cd4a129e2acc3a76d856cec427c5e93dcec1dc28f47cfa0f4ef1e00c2dee24cc, selection sha256:8d4747a05d1213146c97d37a574a457ce8aeee431a3f9f6a9c67c79021d1f437, read set sha256:f18e4fd7eafa09aabd6204d8a41112a4af8892eb03c0b5d7fb9d3bf5f927b3a3, seed 0, harness 0.6.1. realized cost $107.58, which the adaptive layer reads as the cost of its own instrument.
142 events over 5 probes, in 142 sessions and 3865 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. This page names the lock, the selection and the read set the transcript recorded, and it compares none of them with the tree in front of a reader. headwater generate and headwater probe stale name what moved since the recording, and neither one fails for 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 sha256:8d4747a05d1213146c97d37a574a457ce8aeee431a3f9f6a9c67c79021d1f437, and the probes graded below are the probes it was planned over, as this corpus declares them now. A probe added to this corpus after the recording 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. This file names it as provenance and compares it with nothing, and the same holds for the lock and the selection: an edit to a document, a new probe or a moved lock leaves these bytes alone. headwater generate names each result whose recorded lock, read set or selection moved, and headwater probe stale takes the digest and reports which recorded results a change voided. Neither one fails for it.
Graded by grader 0.5.1. A additive run in the scenario-4 arm, on claude-sonnet-5 at served version claude-sonnet-5.
The verdicts
- HW-PROBE-a-session-records-which-obligation-an-evaluation-discharged (sufficiency, expects patched)
session L1-additive-scenario-4-p2-r1: not satisfied —
relation.reciprocity.missingreported over each of the 1 produced artifact; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r10: not satisfied —relation.reciprocity.missingreported over each of the 1 produced artifact; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r11: satisfied — event 32:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r12: satisfied — event 33:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r13: not satisfied —relation.reciprocity.missingreported over each of the 1 produced artifact; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r14: not satisfied —relation.reciprocity.missingreported over each of the 1 produced artifact; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r15: not satisfied —relation.reciprocity.missingreported over each of the 1 produced artifact; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r16: satisfied — event 37:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r17: satisfied — event 38:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r18: not satisfied — the session produced no artifact, so the oracle had nothing to read; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r19: satisfied — event 40:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r2: satisfied — event 41:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r20: satisfied — event 42:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r21: satisfied — event 43:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r22: satisfied — event 44:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r23: not satisfied —relation.reciprocity.missingreported over each of the 1 produced artifact; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r24: satisfied — event 46:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r25: satisfied — event 47:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r26: satisfied — event 48:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r27: satisfied — event 49:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r28: satisfied — event 50:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r29: satisfied — event 51:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r3: not satisfied —relation.reciprocity.missingreported over each of the 1 produced artifact; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r4: satisfied — event 53:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r5: not satisfied —relation.reciprocity.missingreported over each of the 1 produced artifact; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r6: satisfied — event 55:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r7: not satisfied —relation.reciprocity.missingreported over each of the 1 produced artifact; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r8: satisfied — event 57:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server session L1-additive-scenario-4-p2-r9: satisfied — event 58:relation.reciprocity.missingreported nothing overdocs/evaluations/governs-edges-what-an-anchor-reaches-what-covers-it-when-it-ages-and-where-a-session-meets-it.md; no call to a tool of an MCP server - HW-PROBE-a-session-that-changes-a-verb-s-default-reaches-the-verb-s-interface-contract (navigability, expects opened)
session L1-additive-scenario-4-p3-r1: satisfied — event 59, call 3:
Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p3-r1/docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r10: satisfied — event 60, call 34:Bash {"command":"find docs/interfaces -iname \"*route*\"; echo ---; grep -n \"five\\|Five\\|budget\" docs/interfaces/headwater-route.md 2>/dev/null"}; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r11: satisfied — event 61, call 4:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p3-r11/docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r12: satisfied — event 62, call 3:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p3-r12/docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r13: satisfied — event 63, call 2:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p3-r13/docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r14: satisfied — event 64, call 8:Bash {"command":"grep -rn \"five\\|Five\" docs/interfaces/headwater-route.md docs/spec/05-ai-integration.md docs/decisions/0070-the-matched-purposes-take-turns-at-a-route-budget-and-each-pointer-states-what-reached-it.md 2>/dev/null"}; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r15: satisfied — event 65, call 5:Bash {"command":"grep -rn \"pointers: 5\\|pointers, 5\\|budget.pointers\\|Budget::default\\|default_value.*5\\|\\\"5\\\"\" engine/crates/cli/src engine/crates/query/src engine/crates/adapter/src docs/interfaces/headwater-route.md 2>/dev/null"}; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r16: satisfied — event 66, call 4:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p3-r16/docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r17: satisfied — event 67, call 4:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p3-r17/docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r18: satisfied — event 68, call 4:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p3-r18/docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r19: satisfied — event 69, call 20:Bash {"command":"cd /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p3-r19; ls integrations/vscode/test integrations/jetbrains/test | head -30; for f in $(find integrations -name 'route-pointers.jsonl' -o -name 'route-withheld.jsonl' | grep -v node_modules); do echo $f; grep -o '\"withheld[^,}]*' $f | head -2; grep -o '\"path\"' $f | wc -l; done; sed -n 28,34p docs/decisions/0109-*.md | cut -c1-700; sed -n 36,40p docs/decisions/0071-*.md | cut -c1-500; sed -n 29p docs/process/obligations/0224-*.md; sed -n 33p docs/decisions/0064-*.md | grep -oiE \".{100}(default|five).{100}\"; sed -n 35p docs/evaluations/the-three-discovery*.md | grep -oiE \".{100}(default|five|budget).{100}\"; sed -n 28,36p docs/interfaces/headwater-route.md | cut -c1-600; grep -n \"fifth\\|five\\|budget\" docs/reviews/the-sixty-four-restored-help-strings-checked-against-the-binary.md | cut -c1-80"}; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r2: satisfied — event 70, call 4:Read docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r20: satisfied — event 71, call 4:Read docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r21: satisfied — event 72, call 3:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p3-r21/docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r22: satisfied — event 73, call 4:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p3-r22/docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r23: satisfied — event 74, call 8:Bash {"command":"grep -n -i \"five\\|budget\\|pointers\" docs/interfaces/headwater-route.md | head -80"}; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r24: satisfied — event 75, call 42:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p3-r24/docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r25: satisfied — event 76, call 14:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p3-r25/docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r26: satisfied — event 77, call 4:Read docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r27: satisfied — event 78, call 8:Bash {"command":"cd /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p3-r27\necho \"=== docs/interfaces/headwater-route.md ===\"\ngrep -n -i \"five\\|pointer\\|budget\" docs/interfaces/headwater-route.md\necho \"=== docs/decisions/0070 ===\"\ngrep -n -i \"five\" docs/decisions/0070-the-matched-purposes-take-turns-at-a-route-budget-and-each-pointer-states-what-reached-it.md\necho \"=== engine/crates/query/src/mcp.rs ===\"\ngrep -n -i \"five\\|Budget\\|pointers:\" engine/crates/query/src/mcp.rs\necho \"=== engine/crates/cli/tests/route.rs ===\"\ngrep -n -i \"five\\|Budget::default\\|pointers\" engine/crates/cli/tests/route.rs | head -50\n","description":"Grep targeted files for five/pointer/budget references"}; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r28: satisfied — event 79, call 14:Bash {"command":"grep -n -i \"five\\|budget\" /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p3-r28/docs/interfaces/headwater-route.md"}; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r29: satisfied — event 80, call 5:Bash {"command":"grep -rn \"five\" docs/interfaces/headwater-route.md docs/interfaces/headwater-mcp.md docs/interfaces/headwater-probe.md docs/decisions/0070-*.md docs/decisions/0064-*.md docs/evaluations/*.md docs/spec/*.md docs/reviews/the-sixty-four-*.md 2>/dev/null"}; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r3: satisfied — event 81, call 4:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p3-r3/docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r4: satisfied — event 82, call 15:Bash {"command":"cd /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p3-r4\necho \"--- review doc ---\"\ngrep -n \"Five by default\\|ranked pointers\" docs/reviews/the-sixty-four-restored-help-strings-checked-against-the-binary.md\necho \"--- interfaces headwater-route.md ---\"\ngrep -n \"five\\|Five\" docs/interfaces/headwater-route.md\necho \"--- help.rs test ---\"\ngrep -n \"Five by default\\|budget\\|pointers\" engine/crates/cli/tests/help.rs | head -50\necho \"--- spec 05 ---\"\ngrep -n \"five\\|Five\" docs/spec/05-ai-integration.md\n"}; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r5: satisfied — event 83, call 7:Read docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r6: satisfied — event 84, call 4:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p3-r6/docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r7: satisfied — event 85, call 3:Grep docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r8: satisfied — event 86, call 5:Grep docs/interfaces/headwater-route.md; no call to a tool of an MCP server session L1-additive-scenario-4-p3-r9: satisfied — event 87, call 2:Grep docs/interfaces/headwater-route.md; no call to a tool of an MCP server - HW-PROBE-a-session-that-changes-the-recorder-reaches-the-recorder-contract (navigability, expects opened)
session L1-additive-scenario-4-p4-r1: not satisfied — 3 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r10: not satisfied — 7 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r11: not satisfied — 5 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r12: not satisfied — 8 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r13: not satisfied — 7 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r14: not satisfied — 6 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r15: not satisfied — 6 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r16: not satisfied — 5 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r17: not satisfied — 6 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r18: not satisfied — 5 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r19: not satisfied — 9 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r2: not satisfied — 16 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r20: not satisfied — 5 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r21: not satisfied — 5 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r22: not satisfied — 5 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r23: not satisfied — 3 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r24: not satisfied — 5 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r25: not satisfied — 5 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r26: not satisfied — 2 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p4-r27: satisfied — event 107, call 5:
Bash {"command":"grep -n \"no model prose\\|holds no prose\\|prose\" docs/spec/15-the-recorder-contract.md","description":"Show prose-related rules in spec 15"}; no call to a tool of an MCP server session L1-additive-scenario-4-p4-r28: not satisfied — 7 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p4-r29: not satisfied — 5 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p4-r3: not satisfied — 10 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p4-r4: satisfied — event 111, call 6:Bash {"command":"grep -n -i \"prose\\|values a session log omits\\|answer\\b\" \"/mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p4-r4/docs/spec/15-the-recorder-contract.md\" | head -60","description":"Search spec 15 for relevant sections"}; no call to a tool of an MCP server session L1-additive-scenario-4-p4-r5: not satisfied — 7 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p4-r6: not satisfied — 5 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p4-r7: not satisfied — 5 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p4-r8: not satisfied — 2 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p4-r9: not satisfied — 4 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server - HW-PROBE-a-session-that-moves-a-shelf-reaches-the-relocation-guide (discovery, expects opened)
session L1-additive-scenario-4-p5-r1: not satisfied — 54 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
session L1-additive-scenario-4-p5-r10: satisfied — event 118, call 15:
Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p5-r10/docs/how-to/relocate-a-shelf.md; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r11: satisfied — event 119, call 14:Bash {"command":"cd /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p5-r11\nfind docs -iname \"relocate-a-shelf*\"\ncat docs/how-to/relocate-a-shelf.md 2>/dev/null"}; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r12: satisfied — event 120, call 15:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p5-r12/docs/how-to/relocate-a-shelf.md; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r13: satisfied — event 121, call 9:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p5-r13/docs/how-to/relocate-a-shelf.md; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r14: not satisfied — 56 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r15: satisfied — event 123, call 36:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p5-r15/docs/how-to/relocate-a-shelf.md; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r16: satisfied — event 124, call 3:Read ./docs/how-to/relocate-a-shelf.md; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r17: satisfied — event 125, call 41:Bash {"command":"cd /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p5-r17\nfind docs -iname \"*relocate*\"\ncat docs/how-to/relocate-a-shelf.md","description":"Read the existing how-to on relocating a shelf"}; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r18: satisfied — event 126, call 11:Bash {"command":"find docs -iname \"*relocate*\"; echo ---; cat docs/how-to/relocate-a-shelf.md 2>/dev/null"}; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r2: not satisfied — 62 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r20: satisfied — event 128, call 38:Read docs/how-to/relocate-a-shelf.md; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r21: not satisfied — 46 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r22: satisfied — event 130, call 24:Bash {"command":"cat docs/how-to/relocate-a-shelf.md"}; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r23: satisfied — event 131, call 5:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p5-r23/docs/how-to/relocate-a-shelf.md; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r24: not satisfied — 67 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r25: satisfied — event 133, call 31:Bash {"command":"cd /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p5-r25\nfind docs/how-to -iname \"*relocate*\"\ncat docs/how-to/relocate-a-shelf.md 2>/dev/null"}; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r26: not satisfied — 39 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r28: not satisfied — 32 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r29: not satisfied — 42 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r3: not satisfied — 51 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r4: not satisfied — 75 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r5: satisfied — event 139, call 18:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p5-r5/docs/how-to/relocate-a-shelf.md; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r7: not satisfied — 70 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r8: satisfied — event 141, call 17:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p5-r8/docs/how-to/relocate-a-shelf.md; no call to a tool of an MCP server session L1-additive-scenario-4-p5-r9: satisfied — event 142, call 12:Bash {"command":"find . -iname \"*.md\" -path \"*how-to*\" | xargs grep -ln \"relocate\\|move a document\\|mv \" 2>/dev/null | grep -v site/; echo \"---\"; cat docs/how-to/relocate-a-shelf.md 2>/dev/null | head -100"}; no call to a tool of an MCP server - HW-PROBE-an-agent-reaches-the-adjudication-when-the-decision-a-tombstone-links-is-superseded (navigability, expects opened)
session L1-additive-scenario-4-p1-r1: satisfied — event 1, call 4:
Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p1-r1/docs/decisions/0087-no-paragraph-limit-joins-the-language-rule-because-most-paragraphs-past-six-sentences-hold-one-topic.md; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r10: not satisfied — 4 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r11: not satisfied — 2 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r12: satisfied — event 4, call 10:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p1-r12/docs/decisions/0087-no-paragraph-limit-joins-the-language-rule-because-most-paragraphs-past-six-sentences-hold-one-topic.md; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r13: satisfied — event 5, call 6:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p1-r13/docs/decisions/0087-no-paragraph-limit-joins-the-language-rule-because-most-paragraphs-past-six-sentences-hold-one-topic.md; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r14: not satisfied — 1 recorded call, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r15: satisfied — event 7, call 5:Bash {"command":"cat \"/mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p1-r15/docs/decisions/0087-no-paragraph-limit-joins-the-language-rule-because-most-paragraphs-past-six-sentences-hold-one-topic.md\" 2>/dev/null | head -100","description":"Show superseding decision record 0087"}; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r16: satisfied — event 8, call 6:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p1-r16/docs/decisions/0087-no-paragraph-limit-joins-the-language-rule-because-most-paragraphs-past-six-sentences-hold-one-topic.md; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r17: not satisfied — 5 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r18: satisfied — event 10, call 5:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p1-r18/docs/decisions/0087-no-paragraph-limit-joins-the-language-rule-because-most-paragraphs-past-six-sentences-hold-one-topic.md; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r19: satisfied — event 11, call 3:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p1-r19/docs/decisions/0087-no-paragraph-limit-joins-the-language-rule-because-most-paragraphs-past-six-sentences-hold-one-topic.md; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r2: not satisfied — 1 recorded call, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r20: not satisfied — 3 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r21: satisfied — event 14, call 3:Read docs/decisions/0087-no-paragraph-limit-joins-the-language-rule-because-most-paragraphs-past-six-sentences-hold-one-topic.md; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r22: satisfied — event 15, call 3:Read docs/decisions/0087-no-paragraph-limit-joins-the-language-rule-because-most-paragraphs-past-six-sentences-hold-one-topic.md; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r23: satisfied — event 16, call 5:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p1-r23/docs/decisions/0087-no-paragraph-limit-joins-the-language-rule-because-most-paragraphs-past-six-sentences-hold-one-topic.md; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r24: not satisfied — 5 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r25: satisfied — event 18, call 6:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p1-r25/docs/decisions/0087-no-paragraph-limit-joins-the-language-rule-because-most-paragraphs-past-six-sentences-hold-one-topic.md; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r26: satisfied — event 19, call 8:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p1-r26/docs/decisions/0087-no-paragraph-limit-joins-the-language-rule-because-most-paragraphs-past-six-sentences-hold-one-topic.md; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r27: not satisfied — 2 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r28: satisfied — event 21, call 4:Read /mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p1-r28/docs/decisions/0087-no-paragraph-limit-joins-the-language-rule-because-most-paragraphs-past-six-sentences-hold-one-topic.md; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r29: satisfied — event 22, call 6:Bash {"command":"sed -n '1,40p' \"/mnt/build/hw-1777/skills-r1-best/a/ws/L1-additive-scenario-4-p1-r29/docs/decisions/0087-no-paragraph-limit-joins-the-language-rule-because-most-paragraphs-past-six-sentences-hold-one-topic.md\"","description":"Read the superseding decision 0087 for the final outcome"}; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r3: satisfied — event 23, call 3:Read docs/decisions/0087-no-paragraph-limit-joins-the-language-rule-because-most-paragraphs-past-six-sentences-hold-one-topic.md; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r4: not satisfied — 1 recorded call, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r5: not satisfied — 5 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r6: not satisfied — 1 recorded call, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r7: satisfied — event 27, call 7:Read docs/decisions/0087-no-paragraph-limit-joins-the-language-rule-because-most-paragraphs-past-six-sentences-hold-one-topic.md; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r8: not satisfied — 2 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server session L1-additive-scenario-4-p1-r9: not satisfied — 2 recorded calls, and none named any of the 1 document; no call to a tool of an MCP server
142 of 142 sessions that recorded their calls made no call to a tool of an MCP server. 0 sessions recorded no calls.
The rate, and the denominator it is over
81 of 142 graded sessions satisfied their expectation: 57.0%, in a 95% interval of 48.8% to 64.9%. 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.