sisyphi 1.0.8 → 1.0.10

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 (23) hide show
  1. package/dist/daemon.js +13 -8
  2. package/dist/daemon.js.map +1 -1
  3. package/dist/templates/agent-plugin/hooks/debug-user-prompt.sh +15 -0
  4. package/dist/templates/agent-plugin/hooks/operator-user-prompt.sh +14 -0
  5. package/dist/templates/agent-plugin/hooks/review-plan-user-prompt.sh +16 -0
  6. package/dist/templates/agent-plugin/hooks/review-user-prompt.sh +16 -0
  7. package/dist/templates/agent-plugin/hooks/test-spec-user-prompt.sh +14 -0
  8. package/dist/templates/orchestrator-base.md +9 -9
  9. package/dist/templates/orchestrator-impl.md +15 -9
  10. package/dist/templates/orchestrator-plugin/skills/orchestration/SKILL.md +1 -1
  11. package/dist/templates/orchestrator-plugin/skills/orchestration/task-patterns.md +26 -12
  12. package/dist/templates/orchestrator-plugin/skills/orchestration/workflow-examples.md +55 -4
  13. package/package.json +1 -1
  14. package/templates/agent-plugin/hooks/debug-user-prompt.sh +15 -0
  15. package/templates/agent-plugin/hooks/operator-user-prompt.sh +14 -0
  16. package/templates/agent-plugin/hooks/review-plan-user-prompt.sh +16 -0
  17. package/templates/agent-plugin/hooks/review-user-prompt.sh +16 -0
  18. package/templates/agent-plugin/hooks/test-spec-user-prompt.sh +14 -0
  19. package/templates/orchestrator-base.md +9 -9
  20. package/templates/orchestrator-impl.md +15 -9
  21. package/templates/orchestrator-plugin/skills/orchestration/SKILL.md +1 -1
  22. package/templates/orchestrator-plugin/skills/orchestration/task-patterns.md +26 -12
  23. package/templates/orchestrator-plugin/skills/orchestration/workflow-examples.md +55 -4
@@ -78,28 +78,33 @@ Feature with moderate complexity. Requirements may need clarification. Multiple
78
78
  ### Implementation
79
79
  - [ ] Phase 1 — [foundation/types/interfaces]
80
80
  - [ ] Phase 2 — [core logic]
81
+ - [ ] Critique phases 1-2
81
82
  - [ ] Phase 3 — [integration/wiring]
82
-
83
- ### Validation
84
- - [ ] Validate full implementation
83
+ - [ ] Validate — smoketest full feature e2e
85
84
  - [ ] Review implementation
86
85
  ```
87
86
 
87
+ Note: critique and validation are embedded between implementation phases, not deferred to the end. Phase 1 (types) is low-risk and doesn't need its own review, but critique catches issues before Phase 3 builds on them. Validation happens after integration, when all the pieces come together.
88
+
88
89
  ### Cycle plan
89
90
  - **Cycle 1**: Spawn `sisyphus:spec-draft` for spec. Yield. (Human iterates on spec between cycles.)
90
91
  - **Cycle 2**: Spawn `sisyphus:plan` for plan. Yield.
91
92
  - **Cycle 3**: Spawn `sisyphus:review-plan` for review. If fail, respawn plan with issues. Yield.
92
93
  - **Cycle 4**: Spawn `sisyphus:implement` for Phase 1. Yield.
93
- - **Cycle 5**: Spawn `sisyphus:implement` for Phase 2 + `sisyphus:validate` for Phase 1 (parallel if independent). Yield.
94
- - **Cycle 6-8**: Continue phases, validate, review.
94
+ - **Cycle 5**: Spawn `sisyphus:implement` for Phase 2. Phase 1 is types low risk, doesn't need its own validation. Yield.
95
+ - **Cycle 6**: Spawn `sisyphus:review` for critique of phases 1-2. This is the checkpoint before integration builds on top. Yield.
96
+ - **Cycle 7**: Address critique findings + spawn `sisyphus:implement` for Phase 3. Yield.
97
+ - **Cycle 8**: Spawn `sisyphus:validate` for e2e smoketest. Yield.
98
+ - **Cycle 9**: Address validation failures or complete.
95
99
 
96
100
  ### Failure modes
97
101
  - **Spec needs human input**: Mark session as needing human review. Orchestrator notes open questions.
98
102
  - **Plan fails review**: Feed review issues back, respawn planner.
99
- - **Phase fails validation**: Feed specifics back to implement agent for that phase only.
103
+ - **Critique finds issues in foundation**: Fix before starting integration don't build on shaky ground.
104
+ - **Validation fails**: Feed specifics back to implement agent for the failing area.
100
105
 
101
106
  ### Parallelization
102
- Phases without dependencies can run in parallel. Types/interfaces (Phase 1) must complete before implementation phases that consume them.
107
+ Phases without dependencies can run in parallel. Types/interfaces (Phase 1) must complete before implementation phases that consume them. Critique can run alongside detail-planning for the next phase.
103
108
 
104
109
  ---
105
110
 
@@ -119,31 +124,40 @@ Cross-cutting feature, multiple domains, needs team coordination. Uses **progres
119
124
  ### Stage Outline (high-level only — no file-level detail yet)
120
125
  1. [domain A foundation] — no deps — ~N cycles
121
126
  2. [domain B foundation] — no deps — ~N cycles
127
+ → critique stages 1-2 (foundation is low-risk individually, but review before building on it)
122
128
  3. [domain A implementation] — depends on 1 — ~N cycles
123
129
  4. [domain B implementation] — depends on 2 — ~N cycles
130
+ → critique + validate stages 3-4 (core logic, high risk — verify before integration)
124
131
  5. [integration layer] — depends on 3, 4 — ~N cycles
125
- 6. [integration tests] depends on all ~N cycles
132
+ validate end-to-end (integration is where accumulated assumptions break)
133
+ 6. [final review] — depends on all
126
134
 
127
135
  ### Current Stage: [whichever is active]
128
136
  See context/plan-stage-N-{name}.md for detail plan.
129
137
  - [ ] [task-level items from detail plan]
130
138
  ```
131
139
 
140
+ Note: verification checkpoints are embedded in the stage outline, not deferred to a final phase. The level of rigor varies — foundation stages get a light critique, core logic gets critique + validation, integration gets full e2e validation. This is judgment, not formula.
141
+
132
142
  ### Cycle plan
133
143
  - **Cycle 1**: Spawn `sisyphus:spec-draft` for spec. Yield.
134
- - **Cycle 2**: Spawn `sisyphus:plan` for **high-level stage outline only**. Instruction: "Outline stages, dependencies, one-sentence descriptions, cycle estimates. Do not detail any stage no file-level specifics." Spawn `sisyphus:test-spec` for test properties (parallel). Yield.
144
+ - **Cycle 2**: Spawn `sisyphus:plan` for **high-level stage outline only**. Instruction: "Outline stages, dependencies, one-sentence descriptions, cycle estimates. Include verification checkpoints between stages based on risk." Spawn `sisyphus:test-spec` for test properties (parallel). Yield.
135
145
  - **Cycle 3**: Review outline. Spawn `sisyphus:plan` to **detail-plan stage 1 only** (provide outline as context). Output to `context/plan-stage-1-{name}.md`. Yield.
136
146
  - **Cycle 4**: Spawn `sisyphus:implement` for stage 1. If stage 2 is independent, spawn `sisyphus:plan` to detail-plan stage 2 in parallel. Yield.
137
- - **Cycle 5**: Validate stage 1. Spawn `sisyphus:implement` for stage 2 (if detail-planned). Detail-plan stage 3 in parallel if independent. Yield.
138
- - **Cycle 6+**: Continue pattern implement current stage, validate previous, detail-plan next. Each stage follows implement → critique → refine → validate.
147
+ - **Cycle 5**: Spawn `sisyphus:implement` for stage 2 (if detail-planned). Spawn `sisyphus:review` to critique stages 1-2 in parallel — foundation review before core logic builds on it. Detail-plan stage 3 in parallel. Yield.
148
+ - **Cycle 6**: Address critique findings. Spawn `sisyphus:implement` for stage 3. Yield.
149
+ - **Cycle 7**: Spawn `sisyphus:implement` for stage 4. Spawn `sisyphus:review` to critique stage 3 in parallel. Yield.
150
+ - **Cycle 8**: Spawn `sisyphus:validate` for stages 3-4 — core logic checkpoint before integration. Address stage 3 critique. Yield.
151
+ - **Cycle 9+**: Implement integration stage. Validate e2e. Final review.
139
152
 
140
153
  ### Failure modes
141
154
  - **Detail-plan agent can't produce quality output**: The stage is still too large. Break it into sub-stages in the outline and detail-plan each sub-stage individually.
142
155
  - **Integration failures**: Often means contracts between domains don't match. Spawn debug agent targeting the integration seam.
143
156
  - **Stage N implementation invalidates stage N+1 outline**: Update the high-level outline. This is expected — it's why you don't detail-plan everything upfront.
157
+ - **Critique finds issues after multiple stages built on top**: This is the scenario verification checkpoints exist to prevent. If it happens, you waited too long to review — add earlier checkpoints to the roadmap going forward.
144
158
 
145
159
  ### Parallelization
146
- Maximize within the progressive pattern. Independent stages run in parallel. Detail-planning the next stage runs alongside implementing the current one. Foundation stages complete before dependent stages. Integration waits for all domain implementations.
160
+ Maximize within the progressive pattern. Independent stages run in parallel. Detail-planning the next stage runs alongside implementing the current one. Critique and validation agents run alongside the next stage's planning or implementation. Foundation stages complete before dependent stages. Integration waits for all domain implementations.
147
161
 
148
162
  ---
149
163
 
@@ -147,7 +147,7 @@ Agents spawned:
147
147
  against spec at context/spec-rate-limiting.md"
148
148
  ```
149
149
 
150
- ### Cycle 4 — Implement (phases 1+2 parallel)
150
+ ### Cycle 4 — Implement phases 1+2 (parallel, low-risk foundation)
151
151
  ```
152
152
  Plan review: PASS.
153
153
 
@@ -155,16 +155,67 @@ roadmap.md updated (plan review done, starting implementation):
155
155
  - [x] ~~Draft spec~~, [x] ~~Plan~~, [x] ~~Review plan~~
156
156
  - [ ] Implement rate limiting middleware
157
157
  - [ ] Implement rate limit configuration
158
- ...
158
+ - [ ] Critique phases 1-2 — review before integration phase
159
+ - [ ] Implement rate limit headers and error responses
160
+ - [ ] Validate — smoketest rate limiting end-to-end
161
+ - [ ] Final review
159
162
 
160
- Agents spawned:
163
+ Agents spawned (parallel — phases touch different files):
161
164
  implement agent → "Implement Phase 1 from context/plan-rate-limiting.md —
162
165
  rate limiting middleware in src/api/middleware/rate-limit.ts"
163
166
  implement agent → "Implement Phase 2 from context/plan-rate-limiting.md —
164
167
  rate limit configuration in src/config/rate-limits.ts"
165
168
  ```
166
169
 
167
- ### Cycle 5-7Continue phases, validate, review, complete
170
+ ### Cycle 5 — Critique before integration builds on top
171
+ ```
172
+ Both implementation agents complete.
173
+
174
+ Why critique now: Phase 3 (headers/error responses) integrates the middleware and
175
+ config — if the foundation has issues, they'll cascade. Cheaper to catch now.
176
+
177
+ roadmap.md updated:
178
+ - [x] ~~Implement middleware~~, [x] ~~Implement config~~
179
+ - [ ] Critique phases 1-2
180
+ ...
181
+
182
+ Agents spawned:
183
+ review agent → "Review rate limiting middleware and config implementation.
184
+ Focus on: Redis connection handling, sliding window correctness,
185
+ config schema matches what middleware expects."
186
+ ```
187
+
188
+ ### Cycle 6 — Implement phase 3 + address critique
189
+ ```
190
+ Review: 2 findings — middleware doesn't handle Redis connection failure gracefully,
191
+ config schema allows negative rate limits.
192
+
193
+ Agents spawned (parallel):
194
+ implement agent → "Fix review findings in reports/agent-008-final.md for
195
+ rate limiting middleware and config."
196
+ implement agent → "Implement Phase 3 from context/plan-rate-limiting.md —
197
+ rate limit headers and 429 error responses in src/api/middleware/rate-limit.ts"
198
+ ```
199
+
200
+ ### Cycle 7 — Validate end-to-end
201
+ ```
202
+ Phase 3 and fixes complete.
203
+
204
+ Why validate now: all three phases are done and integrated. This is the checkpoint
205
+ before calling it complete — verify it actually works, not just compiles.
206
+
207
+ Agents spawned:
208
+ validate agent → "Verify rate limiting end-to-end: start server, send requests
209
+ exceeding limits, confirm 429 responses with correct Retry-After headers.
210
+ Test per-user isolation, endpoint-specific overrides, Redis failover behavior."
211
+ ```
212
+
213
+ ### Cycle 8 — Complete
214
+ ```
215
+ Validation: PASS. Final review agent confirms no issues.
216
+ Complete — "Added per-user API rate limiting with Redis-backed sliding window,
217
+ configurable per-endpoint limits, and graceful Redis failover."
218
+ ```
168
219
 
169
220
  ---
170
221