cleveragents/cleveragents-core
4
0
Fork 3
Code Issues 6.0k Pull Requests 115 Actions Packages Projects Releases Wiki Activity

docs: architecture corrections cycle 2 — plan revert/resume commands, correction_attempts DDL alignment #5133

Closed
HAL9000 wants to merge 2 commits from spec/architecture-corrections-cycle2 into master
pull from: spec/architecture-corrections-cycle2
merge into: cleveragents:master
Conversation 3 Commits 2 Files Changed 2
+125 -58

2 Commits

Author SHA1 Message Date
HAL9000 3bb842f986 docs: architecture corrections cycle 2 — plan revert/resume commands, correction_attempts DDL alignment
CI / benchmark-publish (pull_request) Has been skipped
Details
CI / lint (pull_request) Failing after 39s
Details
CI / helm (pull_request) Successful in 28s
Details
CI / push-validation (pull_request) Successful in 29s
Details
CI / build (pull_request) Successful in 3m20s
Details
CI / quality (pull_request) Successful in 3m43s
Details
CI / typecheck (pull_request) Successful in 4m2s
Details
CI / security (pull_request) Successful in 4m9s
Details
CI / coverage (pull_request) Has been skipped
Details
CI / benchmark-regression (pull_request) Has been skipped
Details
CI / integration_tests (pull_request) Failing after 4m11s
Details
CI / e2e_tests (pull_request) Successful in 7m15s
Details
CI / unit_tests (pull_request) Successful in 8m35s
Details
CI / docker (pull_request) Has been skipped
Details
CI / status-check (pull_request) Failing after 1s
Details 
Add agents plan revert and agents plan resume to spec command reference (issue #4596)
- Both commands exist in implementation but were absent from spec
- agents plan revert: reverts plan to previous phase (strategize|execute)
- agents plan resume: resumes plan from last checkpoint
- Added to command synopsis and full documentation sections
- Added comparison table distinguishing revert/resume/rollback/correct

Align correction_attempts DDL between early and late spec sections (issue #4911)
- Early DDL: updated PK from attempt_id to correction_attempt_id (matches implementation)
- Early DDL: added mode and guidance fields (matches later DDL and implementation)
- Both DDLs: added original_subtree_snapshot field for correction diff workflow
- Implementation needs Alembic migration to add original_subtree_snapshot column
 2026-04-09 01:23:54 +00:00
HAL9000 1b718e32e6 docs: architecture corrections — invariant precedence, TUI shell safety, sandbox protocol, validation args, ACMS strategy interface
CI / benchmark-publish (pull_request) Has been skipped
Details
CI / lint (pull_request) Failing after 46s
Details
CI / build (pull_request) Successful in 3m17s
Details
CI / helm (pull_request) Successful in 24s
Details
CI / security (pull_request) Successful in 4m8s
Details
CI / quality (pull_request) Successful in 4m12s
Details
CI / push-validation (pull_request) Successful in 15s
Details
CI / typecheck (pull_request) Successful in 4m19s
Details
CI / coverage (pull_request) Has been skipped
Details
CI / benchmark-regression (pull_request) Has been skipped
Details
CI / integration_tests (pull_request) Failing after 6m36s
Details
CI / e2e_tests (pull_request) Successful in 8m9s
Details
CI / unit_tests (pull_request) Successful in 10m53s
Details
CI / docker (pull_request) Has been skipped
Details
CI / status-check (pull_request) Failing after 1s
Details 
Closes #4600 — TUI shell safety: add CRITICAL danger level, update pattern table
Closes #4523 — SandboxStrategyProtocol name, write() return type DiffEntry, registration config keys
Closes #4382 — validation attach --key value format, SemanticEmbeddingStrategy v1 note, SpecStrategyAdapter doc
Closes #4554 — skill YAML: remove skill: wrapper, fix agent_skills_dirs -> agent_skill_folders
Closes #3675 — ACMS reference doc: remove resolved v1 limitations, update strategy interface examples
Fixes invariant precedence pseudocode: plan > action > project > global (was missing action tier)
 2026-04-09 00:49:08 +00:00