@tduyng/pi-devos 0.0.0-stage → 0.13.6

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.
Files changed (138) hide show
  1. package/CHANGELOG.md +1398 -0
  2. package/LICENSE +201 -0
  3. package/README.md +213 -2
  4. package/package.json +45 -4
  5. package/plugins/devos/agents/reader.md +27 -0
  6. package/plugins/devos/agents/verifier.md +181 -0
  7. package/plugins/devos/pi/extension/devos.ts +615 -0
  8. package/plugins/devos/scripts/check-evidence-router.sh +134 -0
  9. package/plugins/devos/scripts/evidence-router.py +252 -0
  10. package/plugins/devos/scripts/gates/check-dod-core.sh +283 -0
  11. package/plugins/devos/scripts/gates/check-dod.sh +38 -0
  12. package/plugins/devos/scripts/gates/check-handoff.sh +215 -0
  13. package/plugins/devos/scripts/gates/check-implementation-boundary.sh +99 -0
  14. package/plugins/devos/scripts/gates/check-port-source.sh +134 -0
  15. package/plugins/devos/scripts/gates/check-report-locations.sh +148 -0
  16. package/plugins/devos/scripts/gates/check-scenario-coverage.sh +102 -0
  17. package/plugins/devos/scripts/gates/gate-check.mjs +204 -0
  18. package/plugins/devos/scripts/gates/phase-gate.sh +208 -0
  19. package/plugins/devos/scripts/gates/plan-template-check.py +147 -0
  20. package/plugins/devos/scripts/gates/run-gate.sh +148 -0
  21. package/plugins/devos/scripts/gates/spec-template-check.py +165 -0
  22. package/plugins/devos/scripts/gates/test-gates.sh +4306 -0
  23. package/plugins/devos/scripts/gates/test-integrity-rubric.sh +152 -0
  24. package/plugins/devos/scripts/gates/test-phase-gate.sh +116 -0
  25. package/plugins/devos/skills/absorb/SKILL.md +75 -0
  26. package/plugins/devos/skills/absorb/references/distill.md +67 -0
  27. package/plugins/devos/skills/absorb/scripts/check-refs.sh +108 -0
  28. package/plugins/devos/skills/absorb/scripts/knowledge.py +258 -0
  29. package/plugins/devos/skills/build/SKILL.md +71 -0
  30. package/plugins/devos/skills/build/references/advanced-routing.md +19 -0
  31. package/plugins/devos/skills/build/references/defect.md +17 -0
  32. package/plugins/devos/skills/build/references/exposure.md +28 -0
  33. package/plugins/devos/skills/build/references/hard.md +39 -0
  34. package/plugins/devos/skills/build/references/refactor.md +66 -0
  35. package/plugins/devos/skills/build/references/routes.md +56 -0
  36. package/plugins/devos/skills/build/references/scale.md +84 -0
  37. package/plugins/devos/skills/build/references/spike.md +19 -0
  38. package/plugins/devos/skills/build/scripts/construct-census.py +204 -0
  39. package/plugins/devos/skills/build/scripts/dep-order.py +297 -0
  40. package/plugins/devos/skills/build/scripts/ledger.sh +228 -0
  41. package/plugins/devos/skills/build/scripts/route.sh +5 -0
  42. package/plugins/devos/skills/build/scripts/secret-scan.sh +121 -0
  43. package/plugins/devos/skills/build/scripts/toolchain-preflight.py +302 -0
  44. package/plugins/devos/skills/design-taste/SKILL.md +55 -0
  45. package/plugins/devos/skills/design-taste/references/tells.md +85 -0
  46. package/plugins/devos/skills/design-taste/scripts/slop-scan.py +305 -0
  47. package/plugins/devos/skills/docs/SKILL.md +22 -0
  48. package/plugins/devos/skills/docs/references/backfill.md +54 -0
  49. package/plugins/devos/skills/docs/references/init.md +43 -0
  50. package/plugins/devos/skills/docs/references/status.md +71 -0
  51. package/plugins/devos/skills/docs/references/update.md +80 -0
  52. package/plugins/devos/skills/explain/SKILL.md +45 -0
  53. package/plugins/devos/skills/explain/references/quiz.md +21 -0
  54. package/plugins/devos/skills/frame/SKILL.md +71 -0
  55. package/plugins/devos/skills/frame/references/boundaries.md +53 -0
  56. package/plugins/devos/skills/frame/references/decompose.md +96 -0
  57. package/plugins/devos/skills/frame/references/impact.md +43 -0
  58. package/plugins/devos/skills/frame/references/interrogate.md +68 -0
  59. package/plugins/devos/skills/frame/scripts/survey.sh +44 -0
  60. package/plugins/devos/skills/implement/SKILL.md +36 -0
  61. package/plugins/devos/skills/implement/references/conditional-plan.md +17 -0
  62. package/plugins/devos/skills/implement/references/debug-loop.md +7 -0
  63. package/plugins/devos/skills/implement/references/dod.md +32 -0
  64. package/plugins/devos/skills/implement/references/implement.md +52 -0
  65. package/plugins/devos/skills/implement/references/manual-gates.md +16 -0
  66. package/plugins/devos/skills/implement/references/plan.md +160 -0
  67. package/plugins/devos/skills/implement/references/resume.md +9 -0
  68. package/plugins/devos/skills/implement/references/scale.md +86 -0
  69. package/plugins/devos/skills/implement/references/verify-close.md +61 -0
  70. package/plugins/devos/skills/implement/references/visual-verification.md +7 -0
  71. package/plugins/devos/skills/implement/scripts/check-converge.sh +147 -0
  72. package/plugins/devos/skills/implement/scripts/equivalence-check.sh +140 -0
  73. package/plugins/devos/skills/implement/scripts/manifest-queue.sh +185 -0
  74. package/plugins/devos/skills/implement/scripts/ops-log.sh +58 -0
  75. package/plugins/devos/skills/implement/scripts/pilot-consistency.py +173 -0
  76. package/plugins/devos/skills/implement/scripts/validate-spec.sh +268 -0
  77. package/plugins/devos/skills/init/SKILL.md +100 -0
  78. package/plugins/devos/skills/init/references/archetype-sections.md +15 -0
  79. package/plugins/devos/skills/init/references/context.md +42 -0
  80. package/plugins/devos/skills/init/references/conventions-template.md +153 -0
  81. package/plugins/devos/skills/init/references/reverse-engineer.md +3 -0
  82. package/plugins/devos/skills/init/references/rule-format.md +58 -0
  83. package/plugins/devos/skills/init/references/shape-model.md +29 -0
  84. package/plugins/devos/skills/init/scripts/detect-archetype.py +461 -0
  85. package/plugins/devos/skills/init/scripts/render-rule.py +114 -0
  86. package/plugins/devos/skills/init/scripts/reverse-engineer.sh +162 -0
  87. package/plugins/devos/skills/learn/SKILL.md +29 -0
  88. package/plugins/devos/skills/learn/references/destinations.md +43 -0
  89. package/plugins/devos/skills/learn/references/scoring.md +26 -0
  90. package/plugins/devos/skills/product/SKILL.md +40 -0
  91. package/plugins/devos/skills/product/references/planner.md +54 -0
  92. package/plugins/devos/skills/review/SKILL.md +77 -0
  93. package/plugins/devos/skills/review/references/gaps.md +26 -0
  94. package/plugins/devos/skills/spec/SKILL.md +58 -0
  95. package/plugins/devos/skills/spec/references/prototype.md +21 -0
  96. package/plugins/devos/skills/spec/references/question-craft.md +47 -0
  97. package/plugins/devos/skills/spec/references/roadmap-link.md +5 -0
  98. package/plugins/devos/skills/spec/references/write-spec.md +143 -0
  99. package/plugins/devos/skills/spec/scripts/clarify.sh +173 -0
  100. package/plugins/devos/skills/status/SKILL.md +67 -0
  101. package/plugins/devos/skills/status/references/cost.md +38 -0
  102. package/plugins/devos/skills/status/scripts/artifact-budgets.sh +88 -0
  103. package/plugins/devos/skills/status/scripts/check-requires.py +556 -0
  104. package/plugins/devos/skills/status/scripts/context-cost.sh +89 -0
  105. package/plugins/devos/skills/status/scripts/convention-usage.sh +184 -0
  106. package/plugins/devos/skills/status/scripts/findings-summary.sh +112 -0
  107. package/plugins/devos/skills/status/scripts/quality-correlation.sh +282 -0
  108. package/plugins/devos/skills/status/scripts/sizing-calibration.sh +71 -0
  109. package/plugins/devos/skills/status/scripts/spec-diff.sh +137 -0
  110. package/plugins/devos/skills/surgical-diff/SKILL.md +23 -0
  111. package/plugins/devos/skills/test-integrity/SKILL.md +26 -0
  112. package/plugins/devos/skills/verification-evidence/SKILL.md +17 -0
  113. package/plugins/devos/workflows/common/backlog-file-structure.md +44 -0
  114. package/plugins/devos/workflows/common/category-file-structure.md +41 -0
  115. package/plugins/devos/workflows/common/feature-line-format.md +10 -0
  116. package/plugins/devos/workflows/common/roadmap-file-structure.md +51 -0
  117. package/plugins/devos/workflows/common/suggest-categories.md +76 -0
  118. package/plugins/devos/workflows/common/suggest-milestones.md +92 -0
  119. package/plugins/devos/workflows/doc/doc-health.md +137 -0
  120. package/plugins/devos/workflows/doc/doc-writer.md +44 -0
  121. package/plugins/devos/workflows/doc/init-documentation.md +196 -0
  122. package/plugins/devos/workflows/doc/retro-documentation.md +304 -0
  123. package/plugins/devos/workflows/doc/templates/library.md +58 -0
  124. package/plugins/devos/workflows/doc/templates/service.md +61 -0
  125. package/plugins/devos/workflows/doc/update-documentation.md +194 -0
  126. package/plugins/devos/workflows/doc/write-documentation.md +137 -0
  127. package/plugins/devos/workflows/product/create-product-mission.md +45 -0
  128. package/plugins/devos/workflows/product/create-product-roadmap.md +106 -0
  129. package/plugins/devos/workflows/product/create-product-tech-stack.md +19 -0
  130. package/plugins/devos/workflows/product/gather-product-info.md +25 -0
  131. package/plugins/devos/workflows/product/resume-detection.md +70 -0
  132. package/plugins/devos/workflows/product/reverse-backlog.md +118 -0
  133. package/plugins/devos/workflows/product/reverse-domain-model.md +86 -0
  134. package/plugins/devos/workflows/product/reverse-product-plan.md +81 -0
  135. package/plugins/devos/workflows/product/reverse-tech-stack.md +73 -0
  136. package/plugins/devos/workflows/product/scan-deep.md +113 -0
  137. package/plugins/devos/workflows/product/scan-git.md +122 -0
  138. package/plugins/devos/workflows/product/scan-light.md +154 -0
package/CHANGELOG.md ADDED
@@ -0,0 +1,1398 @@
1
+ # Changelog
2
+
3
+ ## 0.13.6: 2026-11-11
4
+
5
+ ### Added
6
+ - Add the frame stage and absorbed product knowledge
7
+ - Audit the ask and widen the impact dimensions
8
+ - Name the shape you rejected and why
9
+ - Read the product knowledge before planning
10
+ - Check the knowledge base before quoting it
11
+ - Keep drift signal from drowning in out-of-scope sources
12
+ - Report a spent spike as finished
13
+ - Ship the page-to-code reference check
14
+
15
+ ### Changed
16
+ - Exercise the frame survey script
17
+
18
+ ### Documentation
19
+ - Document absorb and the knowledge files, point init at the manifest script
20
+ - Audit the request for the four things it skipped
21
+ - A frame ends the run that wrote it
22
+ - Say how a project wires itself to a clone
23
+ - A collection page is not the whole collection
24
+ - Say how often the frame may ask
25
+ - Put absorb and frame on the front door
26
+
27
+ ### Fixed
28
+ - Resolve the skill directory outside Claude
29
+ - Count a reimport as nothing when nothing changed
30
+ - A spike folder is finished, not a spec missing its plan
31
+
32
+ ## 0.13.5: 2026-11-10
33
+
34
+ ### Added
35
+ - Hard route and plan gates so smaller models cannot skip phases
36
+ - Route and plan gates on Claude Code, Cursor, Codex, Copilot and opencode
37
+
38
+ ## 0.13.4: 2026-11-08
39
+
40
+ ### Added
41
+ - Restore cursor support as a cursor plugin
42
+ - Block DONE on a scenario the report never records
43
+
44
+ ### Fixed
45
+ - Pass verify on BSD grep, BSD sed, and pi 1.5.0
46
+
47
+ ## 0.13.3: 2026-11-02
48
+
49
+ ### Added
50
+ - Bind the verification report to the commit it judged
51
+ - Name the blind spot and the question a human must answer
52
+ - Name the shared premise, test the returns-nothing case
53
+
54
+ ### Documentation
55
+ - Verify the Pi adapter against pi 1.4.0
56
+ - Fix the Pi skill command typo
57
+ - Name the end-of-turn gate for each harness
58
+ - Refresh the codebase map and conventions to 1744ba4
59
+ - The route entity lists all seven routes
60
+
61
+ ## 0.13.2: 2026-10-26
62
+
63
+ ### Added
64
+ - The verifier runs its real rubric and every path field expands
65
+ - Read the plan and the diff in a fresh context
66
+
67
+ ### Changed
68
+ - Record the port scenarios' honest scores
69
+ - Record the port scenarios' honest scores
70
+ - DevOS ships no agent dispatch and no model choice
71
+ - Drop the System One pass from the evidence router
72
+
73
+ ### Fixed
74
+ - A skill's script paths and adapter tool names resolve on an installed project
75
+
76
+ ## 0.13.1: 2026-10-21
77
+
78
+ ### Added
79
+ - S route offers learn before it stops
80
+ - A scale path for behavior-preserving changes
81
+ - Measure the C route at scale against a real model
82
+ - The port unit order comes from the tree, not from a guess
83
+ - The port's decisions come from a census of the source
84
+ - Measure the port order and the cycle against a real model
85
+ - The port asks what the tree still carries, and the judge has to see a break
86
+ - The port pilots three units and prices the judgment from a measurement
87
+ - A unit ports its test first, and an upgrade reads the release notes it ships with
88
+ - The exposure answer is a scan and a reading, in one reference
89
+ - Measure the port's judgment discipline, and name the port's close
90
+ - A port whose rules do not fit a capsule is a port in phases
91
+ - A phase of a port is a slice of the manifest and a census of that slice
92
+
93
+ ### Changed
94
+ - The amendment's cost is the census count, and the judge rule says it once
95
+ - Record the port scenarios' honest scores
96
+
97
+ ### Documentation
98
+ - The repo map names the artifacts a wide change leaves behind
99
+ - The census has a spec, and two task groups run in order
100
+
101
+ ### Fixed
102
+ - Use a portable in-place sed in two evaluation scripts
103
+ - Ship the Codex marketplace manifest
104
+ - The on-invoke cost ratchet never ran for DevOS's own skills
105
+ - Report the session cost the harness actually loads
106
+ - The template cache must hold every scenario
107
+ - None with a reason is still None, and a port's capsule names a directory
108
+ - The rules reach a worker two ways, and the rulebook stays short
109
+ - The rulebook has to fit the capsule, and the capsule is how a worker gets it
110
+ - A long import chain does not walk off the interpreter's stack
111
+ - The scan prunes vendored trees, and a manifest row with no target is a finding
112
+ - A port's rule is scoped to the file it governs, and it restates at the close
113
+ - A criterion that names a script must see the command, not the reading
114
+ - A criterion sees the whole command, and the ported tree's tests run as they are written
115
+ - A declared guard has to hold before any agent runs
116
+ - A port is one behavior over many files, and C is its route
117
+ - The scale threshold is a file count, because a count can be checked
118
+ - The user asking for the work to be tracked selects the scale path too
119
+
120
+ ## 0.13.0: 2026-10-02
121
+
122
+ ### Added
123
+
124
+ - Ground implementation in local craft
125
+ - Check the plan.md template against the gates, group Phase A into 4 steps
126
+ - Gate unpushed commits on the conventional-commit grammar
127
+ - Check the M spec shape against validate-spec, put it before the L template
128
+ - Add a script census that separates logic from data
129
+ - Teach the doc-drift gate about the check-requires migration
130
+ - Integrate with laya
131
+ - Run laya by default and log what it disagreed with
132
+
133
+ ### Changed
134
+
135
+ - Cover minimal algorithm choice
136
+ - Reduce always-loaded descriptions
137
+ - Remove redundant DevOS prefixes
138
+ - Remove repeated trigger boilerplate
139
+ - Reduce always-loaded descriptions
140
+ - Compress the three phase files to rules without essays
141
+ - Write the functional spec in the interview turn instead of a subagent
142
+ - One question round and one approval gate on the spec pipeline
143
+ - Trim the workspace and quality-row prose
144
+ - Fold the content-gap audit into a synthesis checklist
145
+ - Name a one-page M spec shape and treat 300 lines as a ceiling
146
+ - Move the parallel-marker essay out of the plan template
147
+ - Fold the advanced-routing triggers into the router
148
+ - Put the strange-request rule on the S path
149
+ - Cut the quality-row instruction to the column list
150
+ - List the conventions rules instead of one long paragraph
151
+ - Trim the review brief and the deletion-test restatement
152
+ - Guard that a one-page M spec folder closes the pipeline
153
+ - Group Phase C into 5 steps and scale the close to the change
154
+ - Group Phase B into 4 steps and scale it to the change
155
+ - Put the S test and the S route ahead of the routing table
156
+ - Make planning/notes.md the interview record, not a second spec
157
+ - Fold user stories into each functionality subsection, gate both shapes
158
+ - Drop the duplicated validate-spec step, trigger decisions-input explicitly
159
+ - Write a six-column quality row, gate that the parser counts it
160
+ - Cut the self-trigger argument and two design essays in learn
161
+ - Fold two thin references into write-spec, make the prototype skip decidable
162
+ - State the gate rejections so the planner need not open the scripts
163
+ - Port check-requires to python, 587x faster on --concurrent
164
+ - Add just probe and just census for the unwired audit tools
165
+ - Port slop-scan to python, 14x faster on one file
166
+ - Extract the score heredocs into an importable module
167
+ - Port render-rule to python, 12 processes per rule file
168
+ - Port detect-archetype to python, 14x faster
169
+ - Port check-doc-drift to python, 43x faster
170
+ - Split harness-depth checks out of the per-commit gate
171
+ - Rename the laya pass to System One and make the provider a setting
172
+ - Simplify pi extension
173
+
174
+ ### Documentation
175
+
176
+ - Match the simplified skills, and gate against drift
177
+ - Update docs
178
+
179
+ ### Fixed
180
+
181
+ - Keep process metadata out of implementation output
182
+ - Serialize template cache rebuilds
183
+ - Parallelize independent template builds
184
+ - Bound stubborn tool timeouts
185
+ - Reject blank quality audit fields
186
+ - Reject malformed quality audit components
187
+ - Take findings trend month from the record field
188
+ - Require a first-pass claim for zero-findings credit
189
+ - Report task groups converge cannot identify
190
+ - Keep converge findings on one line
191
+ - Report parallel batches naming an undefined group
192
+ - Compare plan overview count to task groups
193
+ - Accept both markdown bullet forms in convention usage
194
+ - Read task groups with the plan-wide header shape in spec diff
195
+ - Read a size call only from a letter that stands alone in sizing calibration
196
+ - Read the package manager from bun text lockfile in stack facts
197
+ - Say when a capped ledger write dropped content
198
+ - Count an empty frontend file in slop-scan instead of calling it binary
199
+ - Tell a non-repo apart from an empty history in reverse-engineer
200
+ - Read a colon-form task group as a group in the plan gates
201
+ - Compare a colon-form task group in the spec structural diff
202
+ - Match a directory-carrying archetype glob with find -path
203
+ - Search a convention capsule whose spec directory name contains a space
204
+ - Print the quality-correlation route tables in declared S/M/L/C order
205
+ - End a Requires section on any heading, not only a lettered one
206
+ - Reject a rule whose frontmatter never closes
207
+ - Strip the Requires label the same way the reader matches it
208
+ - Gate unresolvable and orphaned references, drop three dead files
209
+ - Move form guidance out of the spec blocks and gate against it leaking back
210
+ - Restore two scenario docs whose edits desynchronised them from the machine criteria
211
+ - Repoint clarify's acceptance-criterion check at the folded spec shape
212
+ - Extend doc drift to the skills, and resolve bare script names
213
+ - Bound the template build lock wait and name the holder
214
+
215
+ ## 0.12.0: 2026-09-15
216
+
217
+ ### Added
218
+
219
+ - Surface human callouts, calibrate findings, drop stale ones
220
+ - Simplify verify step, rewrite plugins installation and remove cursor
221
+ - Restructure plugin tools name
222
+ - Support only opencodeV2, not v1
223
+ - Remove generated tools and scripts
224
+
225
+ ### Changed
226
+
227
+ - Consolidate the three plugins into one devos plugin
228
+ - Reduce to one subagent profile (verifier)
229
+ - Merge spec-write into spec
230
+ - Fold claude-context and domain-model into onboard
231
+ - Merge specs, stats, and diff into status
232
+ - Fold verify-gap into review as --gaps
233
+ - Merge the doc commands into docs and auto-onboard in build
234
+ - Update setup evals
235
+ - Setup devos
236
+ - Rename onboard to init
237
+ - Rename product-plan to product
238
+ - Drop the authority flag, use disable-model-invocation
239
+ - Default to a [spec-name] branch and .worktrees/
240
+ - Rename pi-extension to pi-devos
241
+ - Remove roadmap folder
242
+ - Publish only necessary files
243
+ - Update setup
244
+ - Remove the orphaned post-merge hook and stale symlink comments
245
+ - Wire adopt-check into a recipe and fix a stale AGENTS reference
246
+ - Update pi docs
247
+ - Simplify all skill descriptions
248
+ - Update evals score
249
+
250
+ ### Documentation
251
+
252
+ - Add the DevOS command surface redesign
253
+ - Rewrite the front door with a lifecycle diagram and worked example
254
+ - Drop stale plugin and multi-agent references
255
+ - Record the implemented redesign and its deviations
256
+ - Restore the full route schema and rename stale eval scenarios
257
+ - Remove internal pages and fix support-matrix counts
258
+ - Remove troubleshooting page and its links
259
+ - Shorten configuration and team workflow, drop glossary and situations
260
+ - Remove security, support-matrix, and examples consumers
261
+ - Trim configuration, concepts, and command pages to essentials
262
+ - State why product and docs are optional
263
+ - Remove headless doc
264
+ - Drop per-tool agent references after the generator removal
265
+ - Copilot invokes skills by name, no command wrappers or agent
266
+ - Drop references to removed docs and stale installer comment
267
+ - Refresh the codebase map and the Codex README after the cleanup
268
+ - Pi and Codex no longer ship a verifier agent
269
+ - Drop the Roadmap/Tier/Rule entities and stale conventions
270
+ - Align command syntax, flags, and verifier note with the skills
271
+ - Use GitLab URLs for the Copilot and OpenCode remote installs
272
+ - Drop the last references to the removed roadmaps folder
273
+ - Clear about role member
274
+
275
+ ### Fixed
276
+
277
+ - Check dod score
278
+ - Adapt opencode installer and adapter docs to one plugin
279
+ - Quote the skill description so YAML frontmatter parses
280
+ - Reconcile adapter symlinks so reinstall prunes stale entries
281
+ - Explicite string for skill description
282
+ - Correct plugin paths
283
+ - Close four fail-open gates and correct the install docs
284
+
285
+ ## 0.11.9: 2026-09-12
286
+
287
+ ### Added
288
+
289
+ - Add the H route for fresh-context decomposition
290
+
291
+ ## 0.11.8: 2026-09-10
292
+
293
+ ### Added
294
+
295
+ - Update worktrees rules
296
+
297
+ ### Fixed
298
+
299
+ - Ship the delegate tool by default, discover symlinked agents, map /devos: follow-ups
300
+ - Stop a watchdog subshell from firing the caller's EXIT cleanup
301
+
302
+ ## 0.11.7: 2026-09-09
303
+
304
+ ### Added
305
+
306
+ - Fix missing suggest next step
307
+
308
+ ### Documentation
309
+
310
+ - Add install by project guide
311
+ - Document the per-machine agent override
312
+
313
+ ### Fixed
314
+
315
+ - Wrap /skill: invocations so onboard runs instead of asking
316
+
317
+ ## 0.11.6: 2026-09-05
318
+
319
+ ### Fixed
320
+
321
+ - Probe the gate that exists, make a bare onboard scan instead of asking
322
+ - Catch a stale plugin snapshot at an unchanged version
323
+ - Backfill mandatory conventions sections into an existing file
324
+
325
+ ## 0.11.5: 2026-09-05
326
+
327
+ ### Added
328
+
329
+ - Recommend the smallest option, read the nearest exemplar, detect CI topology
330
+ - Gate the capsule before approval, carry overrides into the spec
331
+ - Fail a plan whose additions answer no failed item
332
+ - Read every runner's checks, detect containers, run checks where the project runs them
333
+ - Poll async work to terminal state, never infer pass from kickoff
334
+
335
+ ### Changed
336
+
337
+ - Improve install command
338
+
339
+ ## 0.11.4: 2026-09-03
340
+
341
+ ### Added
342
+
343
+ - Probe real need and lifecycle states before freezing the spec
344
+ - Falsifiable hypothesis in diagnosis, deletion test for new modules
345
+
346
+ ## 0.11.3: 2026-09-01
347
+
348
+ ### Added
349
+
350
+ - Prewalk-informed cost and closure changes
351
+
352
+ ### Fixed
353
+
354
+ - Rearm pre-commit as the adapter-drift check
355
+
356
+ ## 0.11.2: 2026-09-01
357
+
358
+ ### Changed
359
+
360
+ - Fix skill syntax for pi
361
+ - Regenerate adapters for quoted yaml descriptions
362
+
363
+ ## 0.11.1: 2026-08-31
364
+
365
+ ### Added
366
+
367
+ - Set pi as default, rerun and fix some criteria
368
+ - Baseline cheater probe, 16/32 scenarios pass
369
+ - Ground transcript criteria to filesystem
370
+ - Add word_soup_leak metric, close the last soup leaks
371
+ - Delegate to a chosen model and optional lean pass
372
+
373
+ ### Changed
374
+
375
+ - Golden transcripts for the 8 scenarios without oracles
376
+ - Raise oracle coverage from 7 to 15 scenarios
377
+ - Raise oracle coverage from 7 to 15 scenarios
378
+ - Add oracles for 05b/05c/05d/12/19, fix 19 fixture missing require
379
+ - Add oracles for 03/04/07, fix test-location mismatch
380
+ - Add oracles for 06c/06d, align 06d criterion to scenario
381
+ - Add oracles for 02/06b/20, fix parallel-marker and test-location bugs
382
+ - Complete oracle coverage, all 32 scenarios gated
383
+
384
+ ### Fixed
385
+
386
+ - Criterion and fixture bugs behind 15/17/06g floors
387
+ - Single DoD prefix, name spec error for bare archive path
388
+ - Ellipsis and pasted report path are missing-spec, not missing report
389
+ - Ground 06i, 08, 10, 21 to filesystem artifacts
390
+ - Ground 16 and 08, word_soup_leak stays 0
391
+ - Defeat natural-language pairing cheater
392
+ - Block destructive-cheater hole
393
+ - Close artifact-fabrication cheat
394
+ - 16 phases-reported rejects all wrong phase pairings, not just gamma-complete
395
+ - Reject wrong-direction and negated claims in 08/16
396
+ - Serialize template-cache rebuild with a lock
397
+ - Close toucher gain on 16 via ordered blocked-by chain
398
+ - Enforce must_hold guardrails with a hard exit in devos mode
399
+ - Enforce must_flip discrimination guard with a hard exit in baseline mode
400
+ - Lock in must_flip/must_hold exit-code enforcement with selftests
401
+ - Empty transcript is UNMEASURED, not a precondition score
402
+ - Cover transcript.py's pi schema in the oracle gate
403
+ - Lock multi-turn pi transcript block isolation with selftests
404
+ - Named-evidence resolves ambiguous bare names to any matching file
405
+ - Named-evidence bare-name resolution is case-insensitive
406
+ - Said-near gap is inclusive at the boundary
407
+ - 08 fixture lacked the user stories its prompt demands
408
+ - 03 scores spec-not-overwritten, rebalanced so no-op stays below 50%
409
+ - 18's parallel dispatch was unscored; baseline scored 100%
410
+ - 13-c rename-complete had a corrupted regex (vacuous check)
411
+ - Bare .eval-base in git diff made 10 negative checks vacuous
412
+ - Lock run_with_timeout behavior with selftests
413
+ - 15's done-requires-evidence accepted the stale fixture report
414
+ - 20's plan-recreated was satisfiable by a fabricated empty spec.md
415
+
416
+ ## 0.11.0: 2026-08-26
417
+
418
+ ### Added
419
+
420
+ - Greenfield support - build routes normally on empty repos, onboard skips mapping
421
+ - Preset packs - render-rule accepts directories, presets doc
422
+ - Remove cross-review skill, code-reviewer agent, and all references
423
+ - One-command installer for all 6 tools with dry-run detection
424
+ - Scenario traceability gate - validate-spec cross-checks SC-n coverage between functional spec and plan capsule
425
+ - Implementation notes propagation in capsule; SC-n ids stay out of source comments; refresh measured token rows
426
+ - Rule supersession via frontmatter, render-rule skips dead rules
427
+ - Evolve pass over eval results history
428
+ - Clean-context debug escalation at second non-converging verifier round
429
+ - Must_flip criteria flag with baseline discrimination guard
430
+ - Make pi first principle
431
+ - Self tune
432
+ - Install.sh reports per-tool failures, wire into justfile and README
433
+ - Just update wraps git pull + Claude Code snapshot refresh
434
+
435
+ ### Changed
436
+
437
+ - Improve evals performance
438
+ - First pi eval runs, devos vs baseline on 01-small-fix (75% both)
439
+ - Add check-converge
440
+ - Hold criteria flag
441
+ - Run verification clean pass
442
+ - Prove pi install git:/local-path route against the real pi binary
443
+ - Rename-complete to scenario 13
444
+ - Adapters/pi -> pi-extension, drop the generic adapter wrapper for Pi
445
+ - Diagnostic scenario 19
446
+ - Drop adapters/ wrapper, each tool lives in its native top-level folder
447
+ - Unify symlink installers into data-driven script
448
+ - Add hook-command-path resolution guard
449
+ - Collapse gate shims into data-driven run-gate.sh
450
+ - Add block-probe to check-support-matrix.sh
451
+ - Simplify docs
452
+ - Fix lint scripts
453
+
454
+ ### Documentation
455
+
456
+ - Support-matrix updated
457
+ - Three induced failure modes from pi eval sessions
458
+ - Redgreen authoring protocol, errored-eval-score failure mode
459
+
460
+ ### Fixed
461
+
462
+ - Drop pinned external model IDs from code-reviewer, CLIs track their current default
463
+ - A cwd bug
464
+ - Build router - multi-behavior requests route at least M (pi free-model trial evidence)
465
+ - Errored eval runs score UNMEASURED, never from partial transcripts
466
+ - Scope skill-cost baseline lint to core plugins, not third-party extensions
467
+ - Repair multi-agent install layer
468
+ - Warn gate payload and update docs
469
+
470
+ ## 0.10.16: 2026-08-24
471
+
472
+ ### Added
473
+
474
+ - Optimise token efficient with autoreseearch
475
+
476
+ ## 0.10.15: 2026-08-23
477
+
478
+ ### Added
479
+
480
+ - Ops trail, one JSONL line per phase transition
481
+ - Promote dead ends and behavior conflicts at close, not just findings
482
+
483
+ ## 0.10.14: 2026-08-22
484
+
485
+ ### Added
486
+
487
+ - Register ask_user tool for structured spec-interview questions
488
+ - Acceptance gates, progress-aware pi correction budget
489
+
490
+ ### Changed
491
+
492
+ - Plan approval in conversation, drop harness plan-mode coupling
493
+ - Drop provenance notes from gates work
494
+ - Runner only, drop DoD re-parse of gates.md until a real spec uses one
495
+
496
+ ### Documentation
497
+
498
+ - Full 31-scenario pi run on x-preview-f-free, all at or above floor
499
+
500
+ ### Fixed
501
+
502
+ - Read pi transcripts so transcript criteria score instead of falsely failing
503
+
504
+ ## 0.10.13: 2026-08-21
505
+
506
+ ### Added
507
+
508
+ - Support devos as pi entension
509
+
510
+ ### Documentation
511
+
512
+ - Add Project Context section to AGENTS.md so plain sessions see the artifacts
513
+ - Refresh codebase map to c3f918df
514
+ - Document pi's two-layer CLAUDE_SKILL_DIR resolution
515
+
516
+ ### Fixed
517
+
518
+ - Persist gate correction budget in session entries so reload cannot re-spend it
519
+
520
+ ## 0.10.12: 2026-08-19
521
+
522
+ ### Added
523
+
524
+ - Skill learn can invoke by agent
525
+
526
+ ### Fixed
527
+
528
+ - Recognize it.each/test.each/describe.each as consolidation
529
+ - Reword learn description and update stale token-count docs
530
+
531
+ ## 0.10.11: 2026-08-17
532
+
533
+ ### Added
534
+
535
+ - Rewrite run.sh script, make it argnostic run, no depend to claude
536
+
537
+ ### Changed
538
+
539
+ - Establish per-scenario scoring floors from 2026-08-17 run
540
+ - Run 9 cap-limited scenarios via pi
541
+ - Correct floor values in pi results
542
+ - Update cross-skill lint
543
+
544
+ ## 0.10.10: 2026-08-15
545
+
546
+ ### Added
547
+
548
+ - Enhance guardrails
549
+
550
+ ## 0.10.9: 2026-08-14
551
+
552
+ ### Added
553
+
554
+ - Provenance gates a lesson before durability does
555
+ - Every acceptance criterion carries its own evidence row
556
+ - Findings.md carries what it could not ground
557
+ - Walk back the work the pipeline kept out of your context
558
+
559
+ ## 0.10.8: 2026-08-13
560
+
561
+ ### Added
562
+
563
+ - Simplify opencode adapter
564
+ - Rewrite adapter match opencode2 rules
565
+ - Artifact budget bands enforced in just verify
566
+ - Convention usage report gives conventions.md a prune signal
567
+ - Budget band decides what a write to a capped artifact may do
568
+ - Abandoned approaches get a rejected: decisions entry
569
+
570
+ ### Changed
571
+
572
+ - Domain-model back under its 60-line page budget
573
+ - Drop mine-conventions.sh, unwired and unable to cluster prose
574
+
575
+ ### Documentation
576
+
577
+ - Architecture.md states the map's real 120-line cap
578
+
579
+ ### Fixed
580
+
581
+ - Drop non-devos command wrappers from generated adapter dir
582
+
583
+ ## 0.10.7: 2026-08-11
584
+
585
+ ### Added
586
+
587
+ - Make compatible with claude env
588
+ - Make harness agnostic and fix pi subagent behavior
589
+
590
+ ### Fixed
591
+
592
+ - Install pi better on existing settings
593
+
594
+ ## 0.10.6: 2026-08-11
595
+
596
+ ### Added
597
+
598
+ - Improve evals gates
599
+ - Support piagent and improve stats skill
600
+ - Improve devos for pi agent
601
+ - Fix pi agent scripts
602
+
603
+ ### Changed
604
+
605
+ - Switch to Apache-2.0
606
+ - Fix skills syntax for pi agent
607
+
608
+ ### Documentation
609
+
610
+ - Simplify some page docs
611
+ - Simplify docs, keep shorter
612
+ - Update docs for pi agent
613
+
614
+ ### Fixed
615
+
616
+ - Evals issues
617
+
618
+ ## 0.10.5: 2026-08-10
619
+
620
+ ### Added
621
+
622
+ - Design-taste discipline, anti-slop frontend with a scanner
623
+
624
+ ### Changed
625
+
626
+ - Raise the negative-fixture floor to 10
627
+
628
+ ### Documentation
629
+
630
+ - Route design-taste through the same doc surfaces every skill uses
631
+
632
+ ## 0.10.4: 2026-08-09
633
+
634
+ ### Added
635
+
636
+ - Attach specialist capability instead of absorbing it
637
+
638
+ ### Fixed
639
+
640
+ - Fix promotion script
641
+
642
+ ## 0.10.3: 2026-08-08
643
+
644
+ ### Added
645
+
646
+ - Rerun evals for all skills
647
+
648
+ ### Changed
649
+
650
+ - Close all checkboxes
651
+
652
+ ## 0.10.2: 2026-08-08
653
+
654
+ ### Added
655
+
656
+ - Improve review skills
657
+ - Contradiction is a category, not something the reader notices
658
+ - The C route, plus the fence and grouping it exposed
659
+
660
+ ### Documentation
661
+
662
+ - ROADMAP_023, the framework only knew how to add behavior
663
+ - The correlation report grew two sections, say so
664
+ - The 06-series sweep is deferred, not open work with no owner
665
+ - First real C-route run, cheap and honest about what it found
666
+ - The ∥-dispatch question is answered, weak prescription confirmed on Sonnet too
667
+
668
+ ### Fixed
669
+
670
+ - The 01-05 series ran clean, the 06 series ran into a quota wall
671
+ - C route now hands off to /devos:implement, proven on a real run
672
+
673
+ ## 0.10.1: 2026-08-06
674
+
675
+ ### Added
676
+
677
+ - Hedge words and code blocks are checked, not requested
678
+
679
+ ### Changed
680
+
681
+ - Shape-model.md drops the puppetry, 183 lines to 29
682
+
683
+ ### Documentation
684
+
685
+ - Why spec-functional.md and spec.md stay two files
686
+ - The embed rule and notes.md are load-bearing, not legacy
687
+
688
+ ## 0.10.0: 2026-08-06
689
+
690
+ ### Added
691
+
692
+ - Improve evals
693
+ - Converge on the smallest design before writing the plan
694
+ - A test run that executed nothing is UNVERIFIED, not PASS
695
+ - Gate per-moment load claims (check-claims check 4)
696
+ - The run transcript becomes scoreable evidence
697
+ - The pre-change result is captured, so "pre-existing" is provable
698
+ - Command output is data too, and 06h tests it
699
+
700
+ ### Documentation
701
+
702
+ - Update docs
703
+ - Re-measure the implement phase-load figures (rule 6)
704
+ - One canonical index, duplication cut from every roadmap
705
+ - Compress every roadmap to a changelog, drop unspent-eval tracking
706
+ - 022 closes, with the finding it was not looking for
707
+
708
+ ### Fixed
709
+
710
+ - Improve score calculation
711
+ - .quality.md recorded task-group count as the route
712
+ - The routing analyzer lives where the user's data is
713
+ - The scoring floor moves on the trial median, not the luckiest run
714
+ - The harness could never complete a run, so every floor read 0%
715
+ - Drop --add-dir for the plugin, and the injection report now fires
716
+
717
+ ## 0.9.8: 2026-08-03
718
+
719
+ ### Changed
720
+
721
+ - Move the B and R routes to references/, gate on-invoke cost per skill
722
+
723
+ ### Documentation
724
+
725
+ - Roadmap 021 status, tiers 1-5 are in 0.9.7 and tier 6 is not
726
+
727
+ ## 0.9.7: 2026-08-03
728
+
729
+ ### Added
730
+
731
+ - A defect needs a red-capable command before a hypothesis
732
+ - Give the injection-defense claim a runner, and reach the main agent with it
733
+
734
+ ### Changed
735
+
736
+ - Clean docs
737
+
738
+ ### Documentation
739
+
740
+ - Make docs more clear
741
+ - Add roadmap 21
742
+ - Simplify roadmap 21
743
+ - Roadmap 021, the ops layer rejected and what replaced it
744
+
745
+ ## 0.9.6: 2026-08-01
746
+
747
+ ### Added
748
+
749
+ - Enhance token efficiency
750
+
751
+ ### Documentation
752
+
753
+ - Add docs for devos:learn command
754
+
755
+ ## 0.9.5: 2026-08-01
756
+
757
+ ### Added
758
+
759
+ - A checkable lesson becomes a check, not a bullet
760
+
761
+ ### Documentation
762
+
763
+ - Roadmap 020, the lesson that becomes a check
764
+
765
+ ## 0.9.4: 2026-07-31
766
+
767
+ ### Fixed
768
+
769
+ - Inject the skill into the command, don't ask for it
770
+
771
+ ## 0.9.3: 2026-07-30
772
+
773
+ ### Added
774
+
775
+ - Name the design axes where unspecified has a default
776
+
777
+ ### Documentation
778
+
779
+ - Fix all docs for tiers 40-44 (B route, rule 25, Preview command)
780
+ - Tiers 45 and 46
781
+
782
+ ## 0.9.2: 2026-07-30
783
+
784
+ ### Added
785
+
786
+ - Design direction is decided once, not per feature
787
+
788
+ ## 0.9.1: 2026-07-30
789
+
790
+ ### Added
791
+
792
+ - The app is started and looked at
793
+
794
+ ## 0.9.0: 2026-07-30
795
+
796
+ ### Added
797
+
798
+ - Resolved stack versions reach the capsule
799
+ - Every check the project declares is run and reported by name
800
+ - A defect has a route
801
+
802
+ ### Documentation
803
+
804
+ - Add roadmap 019
805
+ - Tick roadmap 019 tiers 40, 41, 44 and re-measure invoke cost
806
+
807
+ ## 0.8.10: 2026-07-29
808
+
809
+ ### Added
810
+
811
+ - Scenarios carry stable SC-n IDs, coverage becomes mechanical
812
+ - Clickable prototype for UI features, prose is not reviewable
813
+
814
+ ## 0.8.9: 2026-07-29
815
+
816
+ ### Added
817
+
818
+ - Gate the repo against gitlinks and eval-workspace debris
819
+ - Arm just verify as a real pre-commit hook
820
+ - Gate the behavior-delta merge at Verdict: DONE
821
+
822
+ ### Documentation
823
+
824
+ - Add roadmap 018
825
+ - Update ROADMAP_018
826
+
827
+ ### Fixed
828
+
829
+ - Actually stage the env-leak fix from the previous commit
830
+
831
+ ## 0.8.8: 2026-07-28
832
+
833
+ ### Changed
834
+
835
+ - Default verifier and code-reviewer to sonnet
836
+
837
+ ### Documentation
838
+
839
+ - Guidance on picking the model per phase
840
+
841
+ ## 0.8.7: 2026-07-28
842
+
843
+ ### Added
844
+
845
+ - Derive the active spec from the branch, not from mtime
846
+
847
+ ### Changed
848
+
849
+ - Compute context cost in one awk pass
850
+ - Scope check-claims fixture mode to the fixture
851
+
852
+ ### Documentation
853
+
854
+ - Record the 0.8.7 correction pass
855
+
856
+ ## 0.8.6: 2026-07-28
857
+
858
+ ### Added
859
+
860
+ - Derive parallel-safety instead of trusting the marking
861
+ - Gate the per-command invoke-cost table
862
+
863
+ ### Changed
864
+
865
+ - Run the seven deterministic checks concurrently
866
+ - Load each phase when it is reached
867
+ - Pin tiers 31-33, and restore coverage the split dropped
868
+
869
+ ### Documentation
870
+
871
+ - Say where a cheaper model is the right call
872
+ - State that disjointness is now verified, and its limit
873
+ - ROADMAP_017, the framework's own efficiency measured and gated
874
+
875
+ ### Fixed
876
+
877
+ - Surface roster roles the handoff gate silently ignores
878
+
879
+ ## 0.8.5: 2026-07-27
880
+
881
+ ### Added
882
+
883
+ - Build the gate the docs described, wire it on every tool
884
+
885
+ ### Documentation
886
+
887
+ - Consodocs: update architecture flows, add project-type guide
888
+ - Update team-workflow.md with v0.8.4 handoff gate, members.md setup, and role examples
889
+ - List the learn skill in the core component row
890
+ - Clarify team workflow
891
+
892
+ ### Fixed
893
+
894
+ - Adapter DoD shims emit a real decision on both paths
895
+
896
+ ## 0.8.4: 2026-07-27
897
+
898
+ ### Added
899
+
900
+ - Scoring infrastructure and criteria for all 13 scenarios
901
+ - Detection script, conventions template, visual-check marker
902
+ - Quality-correlation, mine-conventions, transitive+timeline views
903
+ - Handoff gate, members template, security documentation
904
+ - Wire handoff gate into copilot and codex Stop hooks
905
+ - Add orphaned worktree with unmerged commits check
906
+ - Embed archetype conventions in plan (Tier 19)
907
+ - Add optional archetype sections to codebase-map template
908
+ - Add archetype-typical verification commands to conventions template
909
+
910
+ ### Changed
911
+
912
+ - Setup scenarios
913
+ - Regenerate handoff adapters after new skill
914
+ - Tick Tier 24 Known Volatility (already in write-spec.md:80)
915
+ - Tick 3 meta notes and ROADMAP_014 correlation item
916
+ - Tick worktree, claims, shims, failure modes, scale items
917
+ - Initialize scoring floor with all 13 scenarios
918
+ - Run 01-small-fix scenario end-to-end, result committed
919
+ - Tick OpenCode eval, scoring floor, failure modes, scale limits
920
+ - Validate Codex/Cursor/Copilot hook configs, close ROADMAP_013 gate-parity
921
+ - Discover hooks for the fail-open sweep instead of listing them
922
+ - Run the 1.0 audit, verdict is fail — 6 pass, 1 partial, 3 fail
923
+ - Record why T21 cannot be closed from inside this repo
924
+
925
+ ### Documentation
926
+
927
+ - Failure modes matrix, scale ceilings, 1.0 audit checklist
928
+ - Update scale-limits.md with real measurements
929
+ - Confirm every inferred marker, fix two wrong ones
930
+ - Induce 7 of 9 rows, two named checks that don't exist
931
+
932
+ ### Fixed
933
+
934
+ - Quality-correlation parsing, adopt-check, brownfield docs, conflict logging
935
+ - Test-integrity rubric counts assertion call sites, not lines
936
+ - Restore original line-matching with consolidation exception
937
+ - Mine-conventions rewrite, detect-archetype JSON, roadmap tick completed items
938
+ - Rename 06c criteria to match scenario name
939
+ - Stop the test-integrity and handoff gates blocking legitimate work
940
+ - Archetype detection never worked and was never invoked
941
+ - Findings-summary counted findings as completed specs
942
+ - Stop mandating a re-verify pass, fix the broken --timeline view
943
+ - Wire --reverse-engineer, correct four more roadmap ticks
944
+
945
+ ## 0.8.3: 2026-07-27
946
+
947
+ ### Added
948
+
949
+ - OpenCode as eval provider in `evals/run.sh` (`EVAL_PROVIDER=opencode`)
950
+ - `evals/with-opencode.sh` auto-installs DevOS plugin into project's `.opencode/` before running OpenCode scenarios
951
+ - Live OpenCode eval results: `evals/results/2026-07-27.md` (3 scenario rows from `opencode/ling-3.0-flash-free`)
952
+ - Per-tool hook I/O shim generation via `HOOKS` config in `scripts/tools.mjs`
953
+ - `generate-adapters.mjs` produces all 6 per-tool DoD and test-integrity shims from config
954
+ - `--check` drift detection verifies generated hook shims match committed shims
955
+
956
+ ### Changed
957
+
958
+ - Tier 12: Wired OpenCode into eval harness with `EVAL_PROVIDER=opencode` env var
959
+ - Tier 12: Updated support-matrix.md OpenCode CI row with new eval plumbing
960
+ - Tier 12: Updated known gaps section to reflect OpenCode eval harness progress
961
+ - Tier 13: Regenerated all 6 hook shims as canonical generator output
962
+ - Regenerated all hook shims from config (tool-specific headers replaced with generated ones)
963
+
964
+ ### Fixed
965
+
966
+ - Generated DoD/test-integrity shims are now generated (rule 21) instead of hand-maintained
967
+
968
+ ## 0.8.2: 2026-07-27
969
+
970
+ ### Added
971
+
972
+ - Onboard this repo — the first real devos/ folder
973
+ - Add time-pressure eval fixture (05d)
974
+ - Add reverse-engineer.sh, git-history-derived spec stubs for brownfield onboarding
975
+
976
+ ### Changed
977
+
978
+ - Complete run-sh-results-suffix, merge behavior delta
979
+ - Fix evals/run.sh results-file truncation
980
+ - Complete pressure-fixture-time-pressure, merge behavior delta
981
+ - Time-pressure eval fixture + test-integrity hook fix
982
+ - Complete reverse-engineer-spec-stubs, merge behavior delta
983
+ - Reverse-engineer.sh spec stubs
984
+
985
+ ### Documentation
986
+
987
+ - Functional spec
988
+ - Functional spec
989
+ - Functional spec
990
+ - Close ROADMAP_016 Tiers 10-11 with real dogfooding receipts
991
+
992
+ ### Fixed
993
+
994
+ - Stop evals/run.sh truncating a same-day results file
995
+ - Test-integrity gate no longer misclassifies devos/specs/*.md as test files
996
+ - Correct reverse-engineer-spec-stubs quality.md route to M (1 task group, not 3+)
997
+
998
+ ## 0.8.1: 2026-07-27
999
+
1000
+ ### Added
1001
+
1002
+ - Give the 24 rules one home in RULES.md
1003
+ - Close the three ungated claim surfaces
1004
+ - Add devos-contract version marker to spec.md and conventions.md
1005
+
1006
+ ### Documentation
1007
+
1008
+ - Resolve all remaining [~] partial-verification items
1009
+
1010
+ ## 0.8.0: 2026-07-27
1011
+
1012
+ ### Added
1013
+
1014
+ - V0.8.0 Tier 1 — data-vs-instruction boundary
1015
+ - V0.8.0 Tier 2 — ambiguity markers in spec-write
1016
+ - V0.8.0 Tier 3 — convention pre-gate in implement
1017
+ - Add scoring rubric for gate-critical scenarios
1018
+
1019
+ ### Documentation
1020
+
1021
+ - Sharpen 015 Conditional — close stale gaps, name next move
1022
+ - Sharpen 015 Conditional — close stale gaps, name next move
1023
+ - Add roadmap 16
1024
+ - Improve roadmap and team roles
1025
+ - Improve roadmap and team roles
1026
+
1027
+ ## 0.7.3: 2026-07-26
1028
+
1029
+ ### Added
1030
+
1031
+ - Declare and derive Requires: dependency edges
1032
+
1033
+ ### Changed
1034
+
1035
+ - Re-deferrals for 014, update 015 conditional table
1036
+ - ROADMAP_014 complete — evals run via opencode/mimo-v2.5-free
1037
+ - ROADMAP_015 precondition updated — ROADMAP_014 complete
1038
+
1039
+ ### Documentation
1040
+
1041
+ - Document /devos:diff
1042
+ - Fix stale function name in evals.md
1043
+ - Add roadmap 015
1044
+ - Sharpen 015 Conditional — close stale gaps, name next move
1045
+
1046
+ ### Fixed
1047
+
1048
+ - Generate correctly path skills for each adapters
1049
+
1050
+ ## 0.7.2: 2026-07-25
1051
+
1052
+ ### Added
1053
+
1054
+ - Rebuild /devos:diff as a real, fixture-tested skill
1055
+
1056
+ ### Fixed
1057
+
1058
+ - Correct false-complete tiers, delete dead/broken scripts
1059
+
1060
+ ## 0.7.1: 2026-07-24
1061
+
1062
+ ### Fixed
1063
+
1064
+ - Resolve 5 regressions found in post-release audit
1065
+
1066
+ ## 0.7.0: 2026-07-24
1067
+
1068
+ ### Added
1069
+
1070
+ - Implement check-adapters drift report, spec quality signal, findings log, diff/timeline/doctor audit, and baseline evals
1071
+ - Add cached templates, test-integrity read-mask prototype, latest eval results
1072
+
1073
+ ### Changed
1074
+
1075
+ - Commit baseline-014.md results for scenarios 01-04
1076
+ - Parallel execution, cached templates, skip-done, configurable max-turns
1077
+
1078
+ ### Documentation
1079
+
1080
+ - Update docs, correct token cost
1081
+ - Add roadmap 014
1082
+ - Update parallel delegation notes and token-efficiency with baseline data
1083
+ - Mark all ROADMAP_014 items as completed
1084
+
1085
+ ## 0.6.6: 2026-07-24
1086
+
1087
+ ### Fixed
1088
+
1089
+ - Make the plugin self-contained so gates survive install
1090
+
1091
+ ## 0.6.5: 2026-07-23
1092
+
1093
+ ### Fixed
1094
+
1095
+ - Symlink devos-product/devos-docs skills on every adapter
1096
+
1097
+ ## 0.6.4: 2026-07-23
1098
+
1099
+ ### Added
1100
+
1101
+ - Extend generator reach to devos-docs
1102
+
1103
+ ## 0.6.3: 2026-07-23
1104
+
1105
+ ### Added
1106
+
1107
+ - Sharper specs before they lock, per-step checks
1108
+
1109
+ ### Changed
1110
+
1111
+ - V0.6.3 sharper specs, need-fit lens, per-step checks (ROADMAP_013)
1112
+
1113
+ ## 0.6.2: 2026-07-23
1114
+
1115
+ ### Added
1116
+
1117
+ - /devos:learn captures and routes durable lessons
1118
+
1119
+ ### Changed
1120
+
1121
+ - V0.6.2 learn loop (ROADMAP_013)
1122
+
1123
+ ## 0.6.1: 2026-07-23
1124
+
1125
+ ### Added
1126
+
1127
+ - Scoped project rules that attach at edit time
1128
+
1129
+ ### Changed
1130
+
1131
+ - V0.6.1 scoped rule layer (ROADMAP_013)
1132
+
1133
+ ## 0.6.0: 2026-07-23
1134
+
1135
+ ### Added
1136
+
1137
+ - Generate agent bodies and command wrappers from source
1138
+ - Deterministic test-integrity floor on Claude Code too
1139
+ - Extend generator reach to the devos-product plugin
1140
+
1141
+ ### Changed
1142
+
1143
+ - V0.6.0 generator, gate parity, devos-product reach (ROADMAP_013)
1144
+
1145
+ ### Documentation
1146
+
1147
+ - Update README
1148
+ - Add roadmap 013
1149
+ - Document the generator contract; reconcile roadmap with what shipped
1150
+
1151
+ ## 0.5.1: 2026-07-22
1152
+
1153
+ ### Added
1154
+
1155
+ - Author gate-under-pressure fixtures, authority and sunk-cost
1156
+
1157
+ ### Changed
1158
+
1159
+ - Baseline run 2026-07-19, single trial, no cap hit
1160
+ - Full suite run with pressure fixtures, no cap hit
1161
+ - Use devos: command syntax to match Claude Code
1162
+
1163
+ ### Documentation
1164
+
1165
+ - Untracked local config does not travel to a worktree, say so
1166
+ - State allowed values per Branching & workspace line, concrete setup examples
1167
+ - Restructure Branching & workspace section, value table and filled example
1168
+ - Configuration reference, every setting with values, defaults, and readers
1169
+ - Catch glossary, implement page, and resuming-work up to branch/workspace behavior
1170
+ - Explicit manual-cleanup section, exact commands, disk cost while a worktree lives
1171
+ - Move 012 deferred items to explicitly rejected, no active requester
1172
+ - Check baseline/parallel eval boxes, findings and fixes documented
1173
+ - Check gate-under-pressure box, mixed findings documented
1174
+
1175
+ ### Fixed
1176
+
1177
+ - Number Phase B pre-steps so resume cannot skip the branch check
1178
+ - Make ∥ dispatch a MUST, not an optional-reading bullet
1179
+ - Generalize pre-steps beyond resume, any urgency framing
1180
+ - Make claude-context more precise
1181
+ - Drop bogus opencode.jsonc plugin registration
1182
+ - Correct hooks.json to nested schema
1183
+
1184
+ ## 0.5.0: 2026-07-18
1185
+
1186
+ ### Added
1187
+
1188
+ - Spec worktrees, L-route isolation by default, declared setup
1189
+ - Flag merged-but-present spec worktrees, gate-tested
1190
+
1191
+ ### Documentation
1192
+
1193
+ - Parallel-work situation page, workspace boundary in architecture and contract
1194
+ - Check 012 tiers 1-2, shipped
1195
+
1196
+ ## 0.4.2: 2026-07-18
1197
+
1198
+ ### Added
1199
+
1200
+ - Spec branch before first edit, default branch is the human's
1201
+
1202
+ ### Changed
1203
+
1204
+ - Add roadmap_012
1205
+
1206
+ ### Documentation
1207
+
1208
+ - Add flow principle
1209
+ - Add ROADMAP_012, a place to work
1210
+ - Branch-per-spec model, recovery entry for commits on the default branch
1211
+
1212
+ ## 0.4.1: 2026-07-17
1213
+
1214
+ ### Added
1215
+
1216
+ - Gate the documented 500-line SKILL.md budget
1217
+ - Pin eval model, add trial count
1218
+ - Add evals recipe, point docs at it
1219
+
1220
+ ### Documentation
1221
+
1222
+ - Explain why conventions.md isn't a best-practices library
1223
+
1224
+ ## 0.4.0: 2026-07-17
1225
+
1226
+ ### Added
1227
+
1228
+ - Behavior delta section in functional specs
1229
+ - Behavior files merged at completion, specs archive stage
1230
+
1231
+ ### Documentation
1232
+
1233
+ - Check 011 tiers 5-7, shipped
1234
+
1235
+ ## 0.3.5: 2026-07-17
1236
+
1237
+ ### Added
1238
+
1239
+ - Scenario discipline, normative wording, removal ritual
1240
+ - Validate-spec checks spec folders mechanically, uniform diagnostic envelope
1241
+ - First-party review skill, calibrated severity, intent-gap work preservation
1242
+
1243
+ ### Documentation
1244
+
1245
+ - ROADMAP_011, the behavior layer
1246
+ - Troubleshooting, reviewing-changes, team-workflow, glossary, agent-contract, worked example, greenfield note
1247
+ - Check 011 tiers 1-4, shipped
1248
+
1249
+ ### Fixed
1250
+
1251
+ - Sync verifier scenario-coverage line into adapter copies
1252
+
1253
+ ## 0.3.4: 2026-07-16
1254
+
1255
+ ### Added
1256
+
1257
+ - Make entry-point skills user-invoked-only again
1258
+ - Gate token claims and command-prefix integrity in just verify
1259
+ - Map freshness signal in build/implement, size budget on the map
1260
+ - Doctor checks adoption health, not just installation
1261
+
1262
+ ### Documentation
1263
+
1264
+ - Update always-loaded claim to measured ~250 tokens
1265
+ - Script-ownership table and ROADMAP_010 (claims become gates)
1266
+
1267
+ ## 0.3.3: 2026-07-16
1268
+
1269
+ ### Documentation
1270
+
1271
+ - Add FAQ answering common objections (pasted ticket, single skill, monorepo, cost)
1272
+ - Pull comparison-table content into FAQ, generalized without naming external tools
1273
+ - Cleanup docs
1274
+ - Add architecture and explain about token usage principle
1275
+ - Fix command prefixes for devos-docs and devos-product modules
1276
+ - Name module entry commands with full prefixes in getting-started
1277
+ - Clarify agent-facing vs human-facing module output
1278
+
1279
+ ## 0.3.2: 2026-07-16
1280
+
1281
+ ### Added
1282
+
1283
+ - Routing collision and fixture-coverage gates, skill-authoring rules (009 tier 4)
1284
+ - Doctor command, version drift, gate scripts, adapter link health (009 tier 5)
1285
+
1286
+ ### Documentation
1287
+
1288
+ - Project-type guidance, wrapper-repo recipe, roadmap forward pointer (009 tier 6)
1289
+
1290
+ ## 0.3.1: 2026-07-16
1291
+
1292
+ ### Added
1293
+
1294
+ - Implementation discipline, clean baseline, named staging, decision self-challenge (009 tier 1)
1295
+ - Verifier depth, multi-line stub detection, blast-radius check order (009 tier 2)
1296
+ - Project-declared verification, proof commands in conventions, human-checked criteria (009 tier 3)
1297
+
1298
+ ### Changed
1299
+
1300
+ - Add roadmap 009
1301
+
1302
+ ## 0.3.0: 2026-07-16
1303
+
1304
+ ### Added
1305
+
1306
+ - Sharpen routing and approval gates (008 tier 1)
1307
+ - Harden the spec capsule (008 tier 2)
1308
+ - Verification protocol, asymmetry, triage, deferred ledger (008 tier 3)
1309
+ - Discipline rationalization tables and anti-summary lint (008 tier 4)
1310
+ - Token-free self-enforcement, gate tests, routing check, release tooling (008 tier 5)
1311
+
1312
+ ### Changed
1313
+
1314
+ - Add comparision tools
1315
+ - Add roadmap 008
1316
+
1317
+ ### Documentation
1318
+
1319
+ - Conventions baseline and 008 doc sync (008 tier 6)
1320
+
1321
+ ## 0.2.0: Parallel execution (token-neutral concurrency)
1322
+
1323
+ - Two admitted shapes of parallelism, both audited to cost zero extra tokens or fewer: concurrent commands inside one agent, and partitioned subagent fan-out where each branch's file set is disjoint from every other branch's.
1324
+ - `/devos:cross-review` now runs its Gemini and Codex reviews concurrently instead of sequentially, identical prompt, identical tokens, wall-clock drops from the sum of both to roughly the slower of the two.
1325
+ - `/devos:cross-review` degrades honestly when neither external CLI is installed: it now says so plainly and points to the harness's own review feature, instead of demanding an install. It never falls back to the harness's own model reviewing its own output, that would rebuild a review the harness already ships and fake the model diversity that is this skill's only reason to exist.
1326
+ - The verifier's quality lens now runs read-only project checks (lint, type-check) concurrently, then build/test checks (which write artifacts or share state) as a batch afterward.
1327
+ - `/devos:implement`'s plan template marks task groups with pairwise-disjoint file sets as parallel-safe (`Group A ∥ Group B, then Group C`), visible at the existing plan-approval gate, no new gate, no new state file. At execution, parallel-safe groups dispatch as concurrent isolated subagents, each reading only the spec capsule and its own group; a merge conflict demotes that group to sequential re-execution rather than attempting a resolution. Never marked below 3 task groups, S/M-sized work stays untouched.
1328
+ - `/devos:onboard --full`'s per-subsystem exploration (already partially in place) is now explicit about applying only to full scans, never to a small incremental re-scan.
1329
+ - `/devos:stats` reports per-branch token cost when a parallel route ran, so the "no more tokens than sequential" claim is checkable per project.
1330
+ - New eval scenarios `06-parallel-vs-sequential` and `06b-conflict-probe` (the gate-breaking test: task groups that share a file must refuse to parallelize) are committed and ready, not yet run against a live model, so `docs/token-efficiency.md` states "not yet measured" rather than inventing a number. Every "faster"/"no more tokens" claim in this release sits next to a real measurement or an honest admission that one doesn't exist yet.
1331
+ - `docs/support-matrix.md` gains a "Parallel delegation" row: native and tuned on Claude Code this release; Codex CLI and OpenCode both have some form of concurrent-subagent capability that isn't adapted yet; Cursor CLI and Copilot CLI have no documented mechanism, so the same prose reads sequentially there by construction.
1332
+ - Retroactively tagged `v0.0.6`, `v0.1.0`, and `v0.1.1`, those releases were versioned in `plugin.json` and documented in this changelog when they shipped, but the git tags themselves were missed at the time.
1333
+
1334
+ ## 0.1.1: Support-matrix convergence
1335
+
1336
+ - 5-column `docs/support-matrix.md` (Claude Code, Cursor CLI, Copilot CLI, Codex CLI, OpenCode); `docs/install/README.md` and the README "Works with" table updated to list all four adapters.
1337
+ - `evals/check-adapters.sh` now covers the two new adapters' duplicated agent bodies (`.toml` for Codex, `.md` for OpenCode), including a TOML-aware body extractor.
1338
+ - All three plugins bumped to `0.1.1`.
1339
+
1340
+ ## 0.1.0: OpenCode adapter (real code, not just symlinks)
1341
+
1342
+ - DevOS now runs on OpenCode, installed via `.opencode/scripts/install-opencode.sh`, symlink-based, `git pull` updates live. Unlike Cursor/Copilot/Codex, this adapter needed real code: OpenCode's skill/command/agent/hook mechanics don't map to translated frontmatter alone (ROADMAP_004 rule 11, one real pattern gap gets a real adapter).
1343
+ - The 11 orchestrators get command wrappers (`.opencode/commands/devos-*.md`, `$ARGUMENTS` supported); the 3 discipline skills and 11 orchestrator skills are symlinked directly (OpenCode's own `skill` tool discovers them); `verifier` and `code-reviewer` are translated to OpenCode's agent format (`mode: subagent`).
1344
+ - Test-integrity is mechanically enforced via a `tool.execute.before` plugin hook (`.opencode/plugins/devos.js`, 90 lines) that denies by throwing.
1345
+ - Definition-of-Done gets a new, honestly-named category, **enforced-via-correction**: OpenCode has no turn-end hook with transcript access, so the plugin reacts to the `session.idle` event, reads the transcript via the SDK client, and injects a correction message rather than blocking outright.
1346
+ - This adapter's hook mechanics were checked against a real, locally installed `opencode` (v1.17.18) and its `@opencode-ai/plugin`/`@opencode-ai/sdk` type definitions, not doc-only. The plugin's own hook logic was unit-tested directly (5 synthetic cases); a full LLM-driven session was not run, to avoid spending API credits without asking.
1347
+ - `docs/install/opencode.md` and an OpenCode column in `docs/support-matrix.md`.
1348
+
1349
+ ## 0.0.6: Codex CLI adapter
1350
+
1351
+ - DevOS now runs on Codex CLI, installed via `devos-codex/scripts/install-codex.sh`, symlink-based, `git pull` updates live.
1352
+ - Re-verifying this roadmap's capability snapshot at implementation time found it stale: Codex CLI hooks and subagents have both shipped since it was written (hooks GA 2026, subagents GA March 2026), upgrading this adapter from the originally-planned "gates degrade to prose" to full gate parity with Copilot CLI.
1353
+ - Both gates are enforced: Definition-of-Done via a `Stop` hook (`transcript_path` in the payload, reusing the same `scripts/gates/check-dod-core.sh` logic as every other adapter), test-integrity via a `PreToolUse` hook denying weakened test edits.
1354
+ - Orchestrators need no command-wrapper layer, Codex reads `SKILL.md` natively via its own skills mechanism, same shape as the Claude Code plugin itself. `verifier` and `code-reviewer` are translated to Codex's TOML custom-agent format (no forced invocation syntax, natural-language delegation only).
1355
+ - No Codex CLI binary was available to test hands-on; every claim is doc-verified and corroborated across multiple independent sources (official docs plus `openai/codex` GitHub issues/PRs), stated plainly in `docs/support-matrix.md`.
1356
+ - `docs/install/codex.md` and a Codex CLI column in `docs/support-matrix.md`.
1357
+
1358
+ ## 0.0.5: Doc fixes following the 0.0.4 release
1359
+
1360
+ - Every existing doc page that showed `/devos:` syntax now says plainly it's Claude Code's, and links to the Cursor/Copilot equivalent and the support matrix, the 0.0.4 release had only added new pages without updating the ones that already existed.
1361
+ - Corrected a claim that had been wrong since 0.0.1: `git pull` alone does **not** update a `claude plugin install`-ed copy of DevOS. `claude plugin install` copies a version-pinned snapshot into `~/.claude/plugins/cache/`; confirmed directly that the cache stays on the old version after a `git pull` until `claude plugin update devos@devos` (+ restart) is run. `README.md`, `docs/concepts.md`, and `docs/install/` now state the real mechanism. Cursor CLI and Copilot CLI are unaffected, their adapters install via symlink, which genuinely is `git pull`-live.
1362
+
1363
+ ## 0.0.4: Cursor CLI and Copilot CLI support
1364
+
1365
+ - DevOS now runs on Cursor CLI and GitHub Copilot CLI, installed via `.cursor/scripts/install-cursor.sh` and `.github/scripts/install-copilot.sh`, symlink-based, `git pull` updates live, same posture as the Claude Code install.
1366
+ - Both adapters get the full orchestrator set (`build`, `spec`, `spec-write`, `implement`, `specs`, `verify-gap`, `cross-review`, `onboard`, `domain-model`, `claude-context`, `stats`) and the two subagents (`verifier`, `code-reviewer`).
1367
+ - Test-integrity is mechanically enforced on both new tools (deterministic weakened-assertion detection, shared logic in `scripts/gates/test-integrity-rubric.sh`).
1368
+ - The Definition-of-Done gate is mechanically enforced on Copilot CLI (same guarantee as Claude Code) and is a documented prose-only discipline on Cursor CLI, because Cursor's stop hook has no transcript access to check against, recorded plainly in `docs/support-matrix.md` rather than shipping a hook that does nothing.
1369
+ - New `docs/support-matrix.md` and `docs/install/` (one page per tool), every capability cell states enforced, prose, or n/a with the reasoning behind it.
1370
+ - New `evals/check-adapters.sh` catches drift between the canonical skill/agent bodies and their per-tool copies.
1371
+ - Every existing doc page that shows `/devos:` syntax (`docs/README.md`, `docs/build.md`, all 9 `docs/commands/*.md` pages, and the `docs/situations/*.md` pages) now says plainly that it's Claude Code's syntax and links to the equivalent Cursor/Copilot invocation and the support matrix, the first pass at this release had only added new pages without updating the existing ones, which still read as Claude-only.
1372
+
1373
+ ## 0.0.3: Harness-neutral core
1374
+
1375
+ - Implementation, spec, domain-model, spec-write, verify-gap, and cross-review skills now describe plan approval, structured Q&A, task tracking, and subagent delegation as behaviors instead of naming Claude-specific tool calls, the same prose now reads correctly on any harness that supports the same primitives.
1376
+ - The project-context workflow writes the same snippet into `AGENTS.md`, so the artifact wiring survives across supported harnesses.
1377
+
1378
+ ## 0.0.2: First tagged release
1379
+
1380
+ - First git tag (`v0.0.1`) for the framework as it stood before this release.
1381
+ - Added a committed eval scenario proving the test-integrity gate holds under a realistic temptation (a failing test with a real source bug, where weakening the assertion is the easy way out).
1382
+ - `CHANGELOG.md` introduced.
1383
+
1384
+ ## 0.0.1: Initial version
1385
+
1386
+ Everything up to this point, never published before now:
1387
+
1388
+ - **Adaptive entry point** (`/devos:build`): sizes any request (S/M/L/R) and routes to the matching amount of process, no ceremony for a small fix, a full spec pipeline only when the task actually needs one.
1389
+ - **Spec pipeline**: `/devos:spec` (interactive Q&A), `/devos:spec-write` (synthesis into a validated functional spec), `/devos:implement` (plan in native plan mode, one approval gate, implement, verify).
1390
+ - **spec.md as a context capsule**: the implementing session reads one file, conventions, domain vocabulary, and prior decisions are embedded in the plan, not re-fetched.
1391
+ - **Enforced gates**: a Definition-of-Done stop hook blocks a false "done" without a verification report; a test-integrity hook denies edits that weaken tests to make them pass.
1392
+ - **Composable verification**: one verifier with two lenses (completeness + quality), findings passed through an adversarial refute pass before being reported; `/devos:verify-gap` for a completeness-only check; `/devos:cross-review` for a second opinion from other AI models on a git diff.
1393
+ - **Project knowledge artifacts**: `conventions.md`, `domain-model.md`, `codebase-map.md`, `decisions.md`, versioned, incrementally updated, shared vocabulary across every skill.
1394
+ - **`/devos:onboard`**: maps an existing codebase into the knowledge artifacts above; incremental re-scan on repeat runs.
1395
+ - **`/devos:stats`**: measures the framework's own always-loaded token cost and per-skill load cost.
1396
+ - **Headless support**: spec-write and verify-gap run non-interactively for CI.
1397
+ - **Optional modules**, installed separately, zero cost unless used: `devos-docs` (documentation system) and `devos-product` (mission, roadmap, backlog, reverse-engineering existing codebases into product docs).
1398
+ - **Distribution**: a local Claude Code plugin marketplace pointed at a git clone, nothing published. **Correction (0.0.4):** "`git pull` updates live" as originally claimed here turned out to be wrong for the installed plugin content itself, confirmed directly, `claude plugin install` copies a version-pinned snapshot; `claude plugin update devos@devos` plus a restart is required. See `docs/install/claude-code.md#update`.