Project.load sites outside runner/ after ADR-0169 #545

New issue

Open

opened 2026-09-09 15:10:37 +00:00 by jared · 0 comments

jared commented

2026-09-09 15:10:37 +00:00

Owner

Context

Follow-up from #539 probe (2026-09-09). ADR-0169 restricts Project.load to open_implementation.rb (entry point, before any bound row exists). #539 enforces this rule for lib/os_sdlc/runner/ files via regression test. Five Project.load sites sit outside runner/ and need caller-chain probe to determine whether each is pre-implementation (explicit target: exemption) or can read the bound row.

Tasks

Probe autocorrect_prepass.rb:95 caller chain; determine whether implementation row exists; if yes, thread it in or pass implementation.project(root); if pre-implementation, keep as exemption with one-line comment citing ADR-0169
Probe lint_command_support.rb caller chain; same decision as autocorrect_prepass.rb
Probe lint_worklist_target.rb caller chain; same decision
Probe project_config_command.rb caller chain; same decision
Probe scope_policy.rb:15 (class method, bare call, no target keyword); note that #537 made this raise ArgumentError if reached without target; determine whether this is dead code or pre-implementation; if pre-implementation, keep as exemption with one-line comment citing ADR-0169 and note #537 context

Acceptance criteria

each of the five files is probed for caller chain and exemption determination recorded in JIRA/Forgejo or a follow-up ticket
any refactored Project.load call passes implementation.project(root) or keeps explicit exemption with citation

Blocking edges

Blocked by: #539

Origin

• Trigger: Agent-initiated from #539 probe findings (code-probe 2026-09-09)
• Improvised this session: none
• Chain: DESIGN ← ADR-0169 ← #534
• Root candidate: #534
• Where: plugins/os-sdlc/lib/os_sdlc/runner and related files
• Session: d233a2f7-977a-419d-99ea-011b6e1e66f8
• Transcript: (from session context)
• Judge: none (agent-filed discovery, skeptic pending)

## Context Follow-up from #539 probe (2026-09-09). ADR-0169 restricts Project.load to open_implementation.rb (entry point, before any bound row exists). #539 enforces this rule for lib/os_sdlc/runner/ files via regression test. Five Project.load sites sit outside runner/ and need caller-chain probe to determine whether each is pre-implementation (explicit target: exemption) or can read the bound row. ## Tasks - [ ] Probe autocorrect_prepass.rb:95 caller chain; determine whether implementation row exists; if yes, thread it in or pass implementation.project(root); if pre-implementation, keep as exemption with one-line comment citing ADR-0169 - [ ] Probe lint_command_support.rb caller chain; same decision as autocorrect_prepass.rb - [ ] Probe lint_worklist_target.rb caller chain; same decision - [ ] Probe project_config_command.rb caller chain; same decision - [ ] Probe scope_policy.rb:15 (class method, bare call, no target keyword); note that #537 made this raise ArgumentError if reached without target; determine whether this is dead code or pre-implementation; if pre-implementation, keep as exemption with one-line comment citing ADR-0169 and note #537 context ## Acceptance criteria - [ ] each of the five files is probed for caller chain and exemption determination recorded in JIRA/Forgejo or a follow-up ticket - [ ] any refactored Project.load call passes implementation.project(root) or keeps explicit exemption with citation ## Blocking edges Blocked by: #539 ## Origin • Trigger: Agent-initiated from #539 probe findings (code-probe 2026-09-09) • Improvised this session: none • Chain: DESIGN ← ADR-0169 ← #534 • Root candidate: #534 • Where: plugins/os-sdlc/lib/os_sdlc/runner and related files • Session: d233a2f7-977a-419d-99ea-011b6e1e66f8 • Transcript: (from session context) • Judge: none (agent-filed discovery, skeptic pending)

jared added the

plugin/os-sdlc

filed-by/agent

labels

2026-09-09 15:10:37 +00:00

jared referenced this issue

2026-09-09 15:44:43 +00:00

#534 C: No Project.load outside open — four call sites #539

jared referenced this issue

2026-09-09 18:53:35 +00:00

#534 F: Handoff facts by convention #542

jared referenced this issue