Commit Graph

Select branches

Hide Pull Requests

audit/resolve-findings

chore/mit-license

claude/test-wsk42i

copilot/configure-github-pages-deployment

docs/detective-court

docs/skill-chapter

feat/explain

fix/skill-frontmatter

main

skill/cross-harness

skill/explain-parity

skill/preflight-ps1

#1

#10

#11

#12

#13

#14

#15

#16

#17

#18

#19

#2

#20

#21

#22

#23

#24

#25

#26

#3

#4

#5

#6

#7

#8

#9

Mono Color

• b18faca993 Merge pull request #25 from bdbrown4/chore/mit-license main benbrownandGitHub 2026-08-04 09:24:21 -04:00
• 856b04dbba chore: relicense GPL-3.0-or-later -> MIT chore/mit-license benbrownandClaude Fable 5 2026-08-04 09:18:21 -04:00
• 10a01acbdf Merge pull request #24 from bdbrown4/docs/detective-court benbrownandGitHub 2026-07-20 13:04:56 -04:00
• 24127bbcb4 docs: name the statistical identity — Bayesian detective, frequentist court docs/detective-court benbrownandClaude Fable 5 2026-07-20 12:52:44 -04:00
• ffc6674e13 Merge pull request #23 from bdbrown4/skill/explain-parity benbrownandGitHub 2026-07-20 10:11:40 -04:00
• 3946438fd6 skill: attempt mode points at explain for the next hypothesis skill/explain-parity benbrownandClaude Fable 5 2026-07-20 10:05:41 -04:00
• c42a46cc92 Merge pull request #22 from bdbrown4/feat/explain benbrownandGitHub 2026-07-20 10:04:08 -04:00
• 444e5294fa explain: post-verdict diagnostics — SPEC §3.7, crate, docs feat/explain benbrownandClaude Fable 5 2026-07-20 09:53:14 -04:00
• a4175c43ba skill: name the terminal case — no launcher at all is a refusal skill/preflight-ps1 benbrownandClaude Fable 5 2026-07-16 09:19:30 -04:00
• 48ef299cec Merge pull request #21 from bdbrown4/skill/preflight-ps1 benbrownandGitHub 2026-07-16 08:44:49 -04:00
• 62df81709d skill: make the sh-vs-ps1 dispatch rule explicit benbrownandClaude Fable 5 2026-07-16 08:32:22 -04:00
• b9fa4932b9 ci: the Windows refusal test failed because the refusal succeeded benbrownandClaude Fable 5 2026-07-16 08:20:49 -04:00
• 0f2020779e skill: cross-OS preflight — ps1 entry point, one CI-enforced contract benbrownandClaude Fable 5 2026-07-16 08:15:43 -04:00
• c5531ed30c Merge pull request #20 from bdbrown4/docs/skill-chapter benbrownandGitHub 2026-07-15 12:24:46 -04:00
• 6e169ab819 docs: bring the book to parity with the skill; PR-gate the book build docs/skill-chapter benbrownandClaude Fable 5 2026-07-15 12:19:49 -04:00
• c2d80f95d8 Merge pull request #19 from bdbrown4/skill/cross-harness benbrownandGitHub 2026-07-15 12:06:18 -04:00
• 7c2177c5c6 skill: executable bit on the mirrored preflight.sh skill/cross-harness benbrownandClaude Fable 5 2026-07-15 11:59:47 -04:00
• 477c5472dd skill: make the peltier skill consumable across agent harnesses benbrownandClaude Fable 5 2026-07-15 11:59:26 -04:00
• ec73e277f1 Merge pull request #18 from bdbrown4/fix/skill-frontmatter benbrownandGitHub 2026-07-14 14:10:42 -04:00
• db5caf0b5d skill: fix unloadable SKILL.md frontmatter, and test it in CI fix/skill-frontmatter benbrownandClaude Opus 4.8 2026-07-14 13:55:14 -04:00
• 314af6959d skill: fix unloadable SKILL.md frontmatter, and test it in CI audit/resolve-findings benbrownandClaude Opus 4.8 2026-07-14 13:55:14 -04:00
• 85eaf43dfe Merge pull request #17 from bdbrown4/audit/resolve-findings benbrownandGitHub 2026-07-14 13:45:47 -04:00
• af3fc3dccd skill: expose the trust layer as a consumable peltier skill benbrownandClaude Opus 4.8 2026-07-14 13:39:38 -04:00
• 899d55f4b6 gates: close the audit findings — enforce the gates the docs claimed benbrownandClaude Opus 4.8 2026-07-14 13:10:14 -04:00
• 21352efe6c Correct project name in README.md benbrownandGitHub 2026-07-12 14:19:24 -04:00
• 1fbead809c Merge pull request #16 from bdbrown4/copilot/configure-github-pages-deployment benbrownandGitHub 2026-07-12 14:17:11 -04:00
• fcf56141bb docs: add configure-pages step to GitHub Pages workflow copilot/configure-github-pages-deployment copilot-swe-agent[bot]andGitHub 2026-07-12 17:45:58 +00:00
• e378097d5b Initial plan copilot-swe-agent[bot]andGitHub 2026-07-12 17:44:40 +00:00
• e55ac2b9c7 Merge pull request #15 from bdbrown4/claude/test-wsk42i benbrownandGitHub 2026-07-12 13:35:47 -04:00
• f2e9036196 gates: set diff-test default-run so cargo run -p diff-test stays unambiguous claude/test-wsk42i Claude 2026-07-12 17:33:05 +00:00
• bc305974d9 docs: rename the working name "hotpath" to peltier.io in README and the book Claude 2026-07-12 17:29:22 +00:00
• 94b3ab9154 Merge pull request #14 from bdbrown4/claude/test-wsk42i benbrownandGitHub 2026-07-11 20:41:05 -04:00
• ff16e67127 phase5: closeout — research-forks case study, docs chapter, roadmap + CLAUDE.md Claude 2026-07-12 00:38:44 +00:00
• 5abeb608e0 phase5: kernel lane (SPEC §13) — FP-tolerance equivalence gate + reference-diff testing Claude 2026-07-12 00:35:31 +00:00
• b7297ad31f phase5: learned class-selection policy (SPEC §13) — rank classes by ledger evidence Claude 2026-07-12 00:24:00 +00:00
• 5d6c259b30 docs: full mdBook site + GitHub Pages deploy; rewrite README as project front page Claude 2026-07-12 00:20:20 +00:00
• a8d1fcce3f ruling: phase0-comrak-002 (mimalloc) ACCEPTED, scoped to the comrak CLI workload Claude 2026-07-12 00:11:45 +00:00
• 42c86a1f6b review: adversarial pass — fix ROI leak for overturned rows + load-gen guards Claude 2026-07-12 00:10:06 +00:00
• d1ef2a0dd6 Merge pull request #13 from bdbrown4/claude/test-wsk42i benbrownandGitHub 2026-07-11 20:00:51 -04:00
• 0ec111663b phase4: COMPLETE — service latency win + mechanical ROI report; roadmap done Claude 2026-07-11 23:48:22 +00:00
• 4cf729e8c1 phase4: mechanical ledger-record helper for the service-latency result Claude 2026-07-11 23:43:08 +00:00
• b12813def8 phase4: service-mode calibration PASS (0/10 A/A false-positive, 10/10 injection detection) Claude 2026-07-11 23:41:57 +00:00
• a78692bbb7 phase4: service-mode latency bench + mechanical ROI report generator Claude 2026-07-11 23:31:58 +00:00
• 3951857cf2 Merge pull request #12 from bdbrown4/claude/test-wsk42i benbrownandGitHub 2026-07-11 19:12:09 -04:00
• 985b85205c phase3: mark COMPLETE in CLAUDE.md — all SPEC §5 exit criteria met Claude 2026-07-11 23:07:48 +00:00
• 03113312ad phase3: cJSON cross-language win verified + audited; robustness fix; case studies Claude 2026-07-11 23:06:59 +00:00
• 832dab3445 phase3: coz integration status; cjson C pipeline proven end-to-end Claude 2026-07-11 18:38:04 +00:00
• b6dab15bf2 phase3: cjson cache-sim profile — float print/parse dominate Claude 2026-07-11 18:35:22 +00:00
• f5f4c10b63 phase3: cjson A/A calibration anchor — PASS Claude 2026-07-11 18:34:46 +00:00
• fd6f963829 phase3: diff-test resolves {out}-templated binary; CI gates cjson Claude 2026-07-11 18:20:08 +00:00
• 014e3b11d1 phase3: generalize trust layer for C/C++; wire cJSON target Claude 2026-07-11 18:17:26 +00:00
• 7f67c3bb02 Merge pull request #11 from bdbrown4/claude/test-wsk42i benbrownandGitHub 2026-07-11 13:57:47 -04:00
• cabac5268d phase2: close out — all SPEC §5 exit criteria met Claude 2026-07-11 17:47:37 +00:00
• 0c4de51a1a agent: phase2-tokei-009 audited — rejected-bench (attempt 20; >=20 bar reached) Claude 2026-07-11 17:12:55 +00:00
• 5797bc3d1b agent: phase2-comrak-011 audited — rejected-bench Claude 2026-07-11 16:59:11 +00:00
• 8ff889b7bd agent: bank + verify phase2-tokei-008, refresh tokei profile Claude 2026-07-11 16:44:05 +00:00
• eaadafc5c9 gates: sanitizer check on the accept path — stop-the-line fix for phase2-comrak-010 Claude 2026-07-11 12:26:30 +00:00
• 206004fbcc agent: comrak diff-fuzzer for the phase2-comrak-010 audit Claude 2026-07-11 12:20:40 +00:00
• 54c63307dc agent: report pipeline death to the polling agent Claude 2026-07-11 12:08:31 +00:00
• c67251614a agent: phase2-tokei-007 audited — rejected-bench Claude 2026-07-11 11:35:52 +00:00
• 3637883c9a agent: phase2-comrak-009 audited — rejected-bench Claude 2026-07-11 11:22:48 +00:00
• f9c3f7f847 agent: phase2-tokei-006 audited — rejected-bench; detached pipeline outlived the capped agent Claude 2026-07-11 11:14:33 +00:00
• c6b25d59f4 agent: phase2-comrak-008 audited — rejected-bench Claude 2026-07-11 02:29:01 +00:00
• b4acf3c3e8 agent: audit-attempt allows /dev/null like the applier; tokei-005 re-audited clean Claude 2026-07-11 02:14:55 +00:00
• 09bcde5bf4 agent: phase2-tokei-005 audited — rejected-bench (class 6, clean null) Claude 2026-07-11 02:13:52 +00:00
• bde70700ca agent: phase2-comrak-007 audited — rejected-bench (+1.0% median, CI includes zero) Claude 2026-07-11 02:04:56 +00:00
• 04daa9122f agent: phase2-tokei-004 audited — rejected-bench Claude 2026-07-11 01:54:35 +00:00
• e28c500942 agent: phase2-comrak-006 audited — rejected-bench (+0.79% median, under bar) Claude 2026-07-11 01:48:00 +00:00
• 82eaeceffb agent: bank + verify phase2-tokei-003, refresh tokei profile Claude 2026-07-10 20:50:21 +00:00
• 3c61c542e8 agent: phase2-tokei-003 accepted by pipeline — manual audit gates in progress Claude 2026-07-10 20:47:53 +00:00
• 1ddc7b3c71 agent: patch-surface fixes from the comrak-005 transcript Claude 2026-07-10 20:36:38 +00:00
• 2386151ddb Merge pull request #10 from bdbrown4/claude/test-wsk42i benbrownandGitHub 2026-07-10 08:25:39 -04:00
• 632e861f7f agent: launcher probes harnessd liveness with a real request Claude 2026-07-10 12:11:42 +00:00
• db6d2dc948 agent: phase2-tokei-002 audited — rejected-gate, golden replay catches live divergence Claude 2026-07-10 05:45:49 +00:00
• 23ab522445 agent: propose_patch ergonomics from the phase2-comrak-004 audit Claude 2026-07-10 05:32:08 +00:00
• 393ca7117f agent: case study — phase2-tokei-001 verified win + audit methodology finding Claude 2026-07-10 05:11:22 +00:00
• abaed3c49d agent: phase2-tokei-001 — first auto-accepted win from the unattended loop Claude 2026-07-10 05:08:16 +00:00
• 189939fe58 agent: transcript progress snapshot Claude 2026-07-10 04:58:29 +00:00
• d49169084f agent: phase2-tokei-001 transcript (in progress) Claude 2026-07-10 04:57:10 +00:00
• 5b198b7091 agent: per-attempt audit script for the Phase 2 grind Claude 2026-07-10 04:52:43 +00:00
• 017491010b bench: cache+branch-sim profiles, hypothesis-granular ledger memory, long-poll read_verdict Claude 2026-07-10 04:49:19 +00:00
• e0c5018008 agent: OS-level isolation for the agent tool boundary (SPEC §10) Claude 2026-07-10 04:43:08 +00:00
• 4eef6a7bd4 Merge pull request #9 from bdbrown4/claude/test-wsk42i benbrownandGitHub 2026-07-09 19:38:59 -04:00
• a7cb94024a agent: cargo fmt for harnessd unknown-op message Claude 2026-07-09 23:34:03 +00:00
• 535821cd48 agent: unattended Phase 2 loop — profile to gated verdict, zero false accepts Claude 2026-07-09 23:29:52 +00:00
• 5be5cad0d3 Merge pull request #8 from bdbrown4/claude/test-wsk42i benbrownandGitHub 2026-07-09 18:39:47 -04:00
• 2813b0b96a phase1: complete — GH-runner 20-session calibration is trust anchor #2 Claude 2026-07-09 22:38:43 +00:00
• dcfcb11bf7 bench: raise CI runner calibration to the full 20-session standard Claude 2026-07-09 22:29:30 +00:00
• 01ac70034c Merge pull request #7 from bdbrown4/claude/test-wsk42i benbrownandGitHub 2026-07-06 20:22:53 -04:00
• 52089c0451 bench: GitHub Actions CI — trust-layer checks, no-net gates, runner calibration Claude 2026-07-07 00:13:21 +00:00
• d7a83c1b3c Merge pull request #6 from bdbrown4/claude/test-wsk42i benbrownandGitHub 2026-07-06 20:08:20 -04:00
• 19e83d878a agent: harnessd IPC — the six-tool surface over line-delimited JSON Claude 2026-07-07 00:03:27 +00:00
• eb32fc24c0 Merge pull request #5 from bdbrown4/claude/test-wsk42i benbrownandGitHub 2026-07-06 19:53:31 -04:00
• baa6c4e967 gates: verdict pipeline — gates + bench + ledger row, pristine-rebuilt baselines Claude 2026-07-06 23:50:05 +00:00
• 5209d8e633 Merge pull request #4 from bdbrown4/claude/test-wsk42i benbrownandGitHub 2026-07-06 19:40:43 -04:00
• 0d29d51f72 gates: diff-test orchestration — corpus pin + tests + golden per target Claude 2026-07-06 23:30:40 +00:00
• 7cd9db4971 Merge pull request #3 from bdbrown4/claude/test-wsk42i benbrownandGitHub 2026-07-06 19:09:09 -04:00
• 73763321c6 bench: automated A/A + regression-injection calibration (Phase 1) Claude 2026-07-06 20:20:40 +00:00
• e69d2c46ed Merge pull request #2 from bdbrown4/claude/test-wsk42i benbrownandGitHub 2026-07-06 16:04:24 -04:00
• 4402775ed2 phase0: exit criteria met — tokei +10.4% median [8.5%, 12.0%] verified Claude 2026-07-06 19:57:05 +00:00