dsh-vibe-math 2.5.0 → 2.6.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.en.md +41 -2
- package/README.md +41 -2
- package/cordis.patch.yml +22 -8
- package/docs/AUDIT-CHECKLIST.md +2 -2
- package/docs/final-paper.md +171 -0
- package/docs/release-notes/RELEASE-NOTES-2.6.0.md +268 -0
- package/docs/test-timing.md +8 -4
- package/package.json +5 -3
- package/prompt-corpus-persona/persona-corpus.json +13 -13
- package/prompt-corpus-persona/persona-corpus.md +49 -21
- package/prompt-corpus-v2/formal-verify-v2.json +12 -0
- package/prompt-corpus-v2/formal-verify-v2.md +144 -53
- package/prompt-corpus-v3/formal-verify-v3.json +36 -0
- package/prompt-corpus-v3/formal-verify-v3.md +405 -78
- package/tests/audit-persona-sensitivity.mjs +60 -7
- package/tests/audit-persona-surface.test.mjs +16 -7
- package/tests/e2e-v5-round2.test.mjs +471 -1
- package/tests/formal-verify-v2.test.mjs +3 -0
- package/tests/formal-verify-v3.test.mjs +4 -0
- package/tests/prompt-v5-integrity.test.mjs +9 -3
- package/tests/selfdrive-v5.mjs +55 -4
- package/vibe-math-v2/agent.cordis.yml +14 -4
- package/vibe-math-v2/vibe-math-v2.js +819 -11
- package/vibe-math-v3/agent.cordis.yml +10 -4
- package/vibe-math-v3/vibe-math-v3.js +800 -11
- package/vibe-math-v4/agent.cordis.yml +12 -4
- package/vibe-math-v4/vibe-math-v4.js +793 -8
- package/vibe-math-v5/agent.cordis.yml +8 -4
- package/vibe-math-v5/vibe-math-v5.js +1116 -8
- package/vibe-math-v5//345/256/236/347/216/260/346/226/271/346/241/210.md +70 -1
|
@@ -61,7 +61,7 @@
|
|
|
61
61
|
Main controls (recommended flow: configure FIRST, then start):
|
|
62
62
|
- vibe_v4_configure {project?, problem?, params?} — create/configure the project (name, problem, params) WITHOUT starting a run; set everything here first.
|
|
63
63
|
- vibe_v4_start {problem?, residentCount?, seedDirections?} — begin the run (spawn residents, brainstorm). If problem was configured, omit it.
|
|
64
|
-
- vibe_v4_set {residentCount, compactThreshold, compactAfterRounds, meetingKeepEvery, maxParallel, activityTimeoutMs, stallAutoMeetingMs, verdictMaxRounds, provider, model, residentPersona, toolAllow, toolDeny, formalVerify, leanCommand, leanArgs, leanTimeoutMs} — tune params (persisted to the settings file). provider/model override the residents' LLM route (empty = they inherit YOUR model/provider); toolAllow/toolDeny are per-resident tool permissions (empty = they inherit all tools). stallAutoMeetingMs is the stalled-group auto-sync-meeting threshold (分级保活 B). formalVerify (off|encourage|require, default off) enables Lean formal verification; leanCommand/leanArgs/leanTimeoutMs configure the toolchain.
|
|
64
|
+
- vibe_v4_set {residentCount, compactThreshold, compactAfterRounds, meetingKeepEvery, maxParallel, activityTimeoutMs, stallAutoMeetingMs, verdictMaxRounds, provider, model, residentPersona, toolAllow, toolDeny, formalVerify, leanCommand, leanArgs, leanTimeoutMs, finalPaper, paperFormat, paperLanguage, paperCompilePdf, paperEditor, paperLatexCommand} — tune params (persisted to the settings file). provider/model override the residents' LLM route (empty = they inherit YOUR model/provider); toolAllow/toolDeny are per-resident tool permissions (empty = they inherit all tools). stallAutoMeetingMs is the stalled-group auto-sync-meeting threshold (分级保活 B). formalVerify (off|encourage|require, default off) enables Lean formal verification; leanCommand/leanArgs/leanTimeoutMs configure the toolchain. The FINAL PAPER is tuned by finalPaper (true|false, default true), paperFormat (both|md|tex), paperLanguage (zh|en), paperCompilePdf (true|false), paperEditor (office|resident:<id>, default office = the human/assistant side finalises) and paperLatexCommand.
|
|
65
65
|
- vibe_v4_resume / vibe_v4_pause / vibe_v4_abort / vibe_v4_status / vibe_v4_report.
|
|
66
66
|
- vibe_v4_message {to|all, content} — inject a message to a resident (human/assistant intervention).
|
|
67
67
|
- vibe_v4_meeting {agenda} — force a meeting.
|
|
@@ -72,7 +72,11 @@
|
|
|
72
72
|
- vibe_v4_formal_report — human-readable Lean formal-verification mirror (mode, Lean-passed
|
|
73
73
|
objects, recorded blockers, formalization TODO, library paths).
|
|
74
74
|
- vibe_v4_prompts {which: brainstorm|normal|heartbeat|meeting|verify|coreRules, member?, target?, stage?} — read the exact prompt text a resident would receive (prompt auditing; prompt text is the product).
|
|
75
|
-
A /v4 slash command mirrors the main controls (configure|start [problem]|resume|pause|abort|status|report|message <to|all> <content>|meeting|members|add|remove|set <k=v>...).
|
|
75
|
+
A /v4 slash command mirrors the main controls (configure|start [problem]|resume|pause|abort|status|report|message <to|all> <content>|meeting|paper [lang=zh|en] [format=md|tex|both] [editor=office|resident:<id>] [force]|members|add|remove|set <k=v>...).
|
|
76
|
+
Final paper: after the unanimous stop vote the run enters a PAPER phase — every resident writes their part, each
|
|
77
|
+
cross-reviews at least one other part, and the paperEditor finalises (office = the human/assistant side, or one named
|
|
78
|
+
resident); only then is the run marked complete. finalPaper=false skips the phase; /v4 paper [lang=zh|en]
|
|
79
|
+
[format=md|tex|both] [editor=office|resident:<id>] [force] starts or re-runs it by hand.
|
|
76
80
|
|
|
77
81
|
TRUST RULE: only Verified/ (and Propos/ entries marked 已验证·真/假) are absolutely
|
|
78
82
|
trustworthy; everything else — unverified resident claims, Progress/, Method claims —
|
|
@@ -153,7 +157,7 @@
|
|
|
153
157
|
Main controls (recommended flow: configure FIRST, then start):
|
|
154
158
|
- vibe_v4_configure {project?, problem?, params?} — create/configure the project (name, problem, params) WITHOUT starting a run; set everything here first.
|
|
155
159
|
- vibe_v4_start {problem?, residentCount?, seedDirections?} — begin the run (spawn residents, brainstorm). If problem was configured, omit it.
|
|
156
|
-
- vibe_v4_set {residentCount, compactThreshold, compactAfterRounds, meetingKeepEvery, maxParallel, activityTimeoutMs, stallAutoMeetingMs, verdictMaxRounds, provider, model, residentPersona, toolAllow, toolDeny, formalVerify, leanCommand, leanArgs, leanTimeoutMs} — tune params (persisted to the settings file). provider/model override the residents' LLM route (empty = they inherit YOUR model/provider); toolAllow/toolDeny are per-resident tool permissions (empty = they inherit all tools). stallAutoMeetingMs is the stalled-group auto-sync-meeting threshold (分级保活 B). formalVerify (off|encourage|require, default off) enables Lean formal verification; leanCommand/leanArgs/leanTimeoutMs configure the toolchain.
|
|
160
|
+
- vibe_v4_set {residentCount, compactThreshold, compactAfterRounds, meetingKeepEvery, maxParallel, activityTimeoutMs, stallAutoMeetingMs, verdictMaxRounds, provider, model, residentPersona, toolAllow, toolDeny, formalVerify, leanCommand, leanArgs, leanTimeoutMs, finalPaper, paperFormat, paperLanguage, paperCompilePdf, paperEditor, paperLatexCommand} — tune params (persisted to the settings file). provider/model override the residents' LLM route (empty = they inherit YOUR model/provider); toolAllow/toolDeny are per-resident tool permissions (empty = they inherit all tools). stallAutoMeetingMs is the stalled-group auto-sync-meeting threshold (分级保活 B). formalVerify (off|encourage|require, default off) enables Lean formal verification; leanCommand/leanArgs/leanTimeoutMs configure the toolchain. The FINAL PAPER is tuned by finalPaper (true|false, default true), paperFormat (both|md|tex), paperLanguage (zh|en), paperCompilePdf (true|false), paperEditor (office|resident:<id>, default office = the human/assistant side finalises) and paperLatexCommand.
|
|
157
161
|
- vibe_v4_resume / vibe_v4_pause / vibe_v4_abort / vibe_v4_status / vibe_v4_report.
|
|
158
162
|
- vibe_v4_message {to|all, content} — inject a message to a resident (human/assistant intervention).
|
|
159
163
|
- vibe_v4_meeting {agenda} — force a meeting.
|
|
@@ -164,7 +168,11 @@
|
|
|
164
168
|
- vibe_v4_formal_report — human-readable Lean formal-verification mirror (mode, Lean-passed
|
|
165
169
|
objects, recorded blockers, formalization TODO, library paths).
|
|
166
170
|
- vibe_v4_prompts {which: brainstorm|normal|heartbeat|meeting|verify|coreRules, member?, target?, stage?} — read the exact prompt text a resident would receive (prompt auditing; prompt text is the product).
|
|
167
|
-
A /v4 slash command mirrors the main controls (configure|start [problem]|resume|pause|abort|status|report|message <to|all> <content>|meeting|members|add|remove|set <k=v>...).
|
|
171
|
+
A /v4 slash command mirrors the main controls (configure|start [problem]|resume|pause|abort|status|report|message <to|all> <content>|meeting|paper [lang=zh|en] [format=md|tex|both] [editor=office|resident:<id>] [force]|members|add|remove|set <k=v>...).
|
|
172
|
+
Final paper: after the unanimous stop vote the run enters a PAPER phase — every resident writes their part, each
|
|
173
|
+
cross-reviews at least one other part, and the paperEditor finalises (office = the human/assistant side, or one named
|
|
174
|
+
resident); only then is the run marked complete. finalPaper=false skips the phase; /v4 paper [lang=zh|en]
|
|
175
|
+
[format=md|tex|both] [editor=office|resident:<id>] [force] starts or re-runs it by hand.
|
|
168
176
|
|
|
169
177
|
TRUST RULE: only Verified/ (and Propos/ entries marked 已验证·真/假) are absolutely
|
|
170
178
|
trustworthy; everything else — unverified resident claims, Progress/, Method claims —
|