fix(concurrency): add thread safety to InvariantService #11066

Closed
HAL9000 wants to merge 15 commits from agents/final-working into master

15 Commits

Author SHA1 Message Date
clever-agent 7d3715bd58 build: restricted bash to durther prevent force merges or sudo escalation
CI / benchmark-publish (pull_request) Has been skipped
CI / build (pull_request) Successful in 1m24s
CI / lint (pull_request) Successful in 1m38s
CI / quality (pull_request) Successful in 1m39s
CI / push-validation (pull_request) Successful in 48s
CI / benchmark-regression (pull_request) Failing after 1m29s
CI / helm (pull_request) Successful in 57s
CI / typecheck (pull_request) Successful in 1m58s
CI / security (pull_request) Successful in 2m1s
CI / e2e_tests (pull_request) Successful in 5m57s
CI / integration_tests (pull_request) Successful in 6m56s
CI / unit_tests (pull_request) Failing after 9m12s
CI / docker (pull_request) Has been skipped
CI / coverage (pull_request) Has been skipped
CI / status-check (pull_request) Failing after 3s
2026-05-07 23:17:32 -04:00
clever-agent d53cdc3607 build: Had to re-restrict the permissions on the agents, they were misbehaving 2026-05-07 21:08:31 -04:00
clever-agent 5d02a7c957 Fixed a bad/old escalation table 2026-05-07 18:10:43 -04:00
HAL9000 b9edc377d8 docs(spec): clarify layer boundary DI exception, ULID scope, TUI/ACMS gap-fill [AUTO-ARCH-1]
Add targeted clarifications to docs/specification.md to fill identified gaps:

1. Layer boundary DI Container Exception (Cross-Milestone Architectural Invariants)
2. ULID Scope Clarification - domain vs internal identifiers
3. ACMS Pipeline Protocol Contracts with storage tiers and budget protocol
4. TUI Component Interfaces with verifiable checks

Co-authored-by: CleverAgents Bot <bot@cleveragents.com>

ISSUES CLOSED: #10451
2026-05-07 10:43:39 +00:00
clever-agent df04392ebb build: testing agents with allow first permissions 2026-05-06 22:36:23 -04:00
clever-agent 8758455aaa build: fixed the tiers to now only use qwen and kimi 2026-05-05 23:27:43 -04:00
clever-agent 581b59f070 build: improved the startup script for auto-agents 2026-05-05 20:19:24 -04:00
clever-agent 3da61ba891 build: opened up echo and cat perms to help smooth over some tool calls 2026-05-05 20:11:19 -04:00
clever-agent b75feaf7a3 build: changed permission on auto-agents script 2026-05-05 18:56:03 -04:00
clever-agent 7dda57b054 build: added deny permission for sudo calls 2026-05-05 18:54:42 -04:00
clever-agent fc5a8a9e6d build: implement explicit-vs-fetched variable propagation across all agents
All agents now track which variables were explicitly present in their prompt
versus fetched from environment variables or git remote. When constructing
subagent prompts, only explicitly-present variables are included. Fetched
variables are omitted, allowing each subagent to fetch them independently.

This prevents credentials and other fetched values from being garbled as they
propagate through multiple LLM prompt layers.

Affected agents:
- auto-agents (primary orchestrator)
- implementation-supervisor, pr-merge-supervisor, pr-review-supervisor
- supervisor (generic)
- implementation-worker, pr-merge-worker, pr-review-worker
- task-implementor, tier-dispatcher
- work-group-util, git-clone-util, git-push-util, git-checkout-util
2026-05-05 22:22:09 +00:00
clever-agent f8f74a56d9 build: added critical rule clearly stating no question asking 2026-05-05 16:54:25 -04:00
clever-agent 02e7cb39f6 build: improved some of the standard bash permissions to access fetchable variables at every level 2026-05-05 16:32:01 -04:00
clever-agent e1bfc970a7 build: Fixed models being used to align with new endpoints 2026-05-05 16:18:33 -04:00
clever-agent e16181a7ec build: pulled in new endpoints and fixing permission from experimental branch 2026-05-05 16:12:53 -04:00