1.9 KiB
1.9 KiB
PROGRESS.md
Current State
- Active objective: project guidance refresh
- Active owner: root agent
- Current status: implementation and static verification complete
- Design:
docs/superpowers/specs/2026-08-18-project-guidance-refresh-design.md - Implementation plan:
docs/superpowers/plans/2026-08-18-project-guidance-refresh.md
Completed
- Audited the active
AGENTS.mdchain and current project documents. - Confirmed the Abaqus manual navigation contract from
INDEX_MAP.mdandINDEX.md. - Confirmed that external CSV comparison remains the solver-result verification path.
- Inventoried remaining versioned files and references for the retired feature.
- Updated the project guidance so FEM theory uses
fem-theory-queryand Abaqus manual facts useINDEX_MAP.md→INDEX.md→ everysource_rangesspan. - Updated the relevant custom agents and skills with mandatory source-routing rules.
- Removed retired project-specific feature documents, phases, Fortran sources, and tests.
- Preserved user-supplied external displacement/stress CSV plus provenance as the Abaqus result-validation contract.
In Progress
- None.
Blocked
- None.
Last Verification
- No harness skill or test command was run, per user instruction.
git diff --check: clean.- Static source-routing and retired-reference scans: clean.
- Agent TOML prompt delimiters and
model_reasoning_effort = "extra high": valid. - Skill frontmatter and required source-routing markers: valid.
- Manual inventory:
INDEX_MAP.md,INDEX.md, and 69 source chunks present; all 115 map entries resolve to the expectedINDEX.mdheading line. - Active
AGENTS.mdchain: root file only, 9,624 bytes, below the 32 KiB instruction limit.
Next Agent Checklist
- Read
AGENTS.md,PLAN.md,PROGRESS.md, andWORKNOTE.md. - When the user supplies externally generated CSV/provenance, validate it against the approved schema and tolerances without running Abaqus locally.