add pdftomd

.codex/commands/plan-mineru-integration.md

@@ -0,0 +1,28 @@
+---
+description: Plan the direct local MinerU CLI adapter and failure/reporting behavior
+argument-hint: [integration-scope]
+allowed-tools: [Read, Glob, Grep, Bash, WebFetch, Edit]
+---
+
+# /plan-mineru-integration
+
+Plan the future implementation shape for the MinerU adapter without writing converter code.
+
+## Arguments
+
+The user invoked this command with: $ARGUMENTS
+
+## Workflow
+
+1. Read `PLAN.md`, `PROGRESS.md`, `PRD.md`, and `ARCHITECTURE.md`.
+2. Verify any MinerU CLI facts that may have changed before changing docs.
+3. Define the smallest adapter contract for command construction, working directories, outputs, stdout/stderr capture, exit handling, warnings, and provenance.
+4. Ensure failure behavior is explicit: no silent fallback and no alternate engine route.
+5. Identify mocked-output tests and optional MinerU-dependent checks.
+6. Update `PLAN.md` only if implementation sequencing changes; update `PROGRESS.md` after the planning work.
+
+## Guardrails
+
+- Do not implement program code during planning.
+- Do not introduce runtime engine selection or cloud-compatible endpoints.
+- Keep GPU limitations and CPU messaging explicit for GTX 1070 Ti 8GB.

.codex/commands/plan-quality-evaluation.md

@@ -0,0 +1,27 @@
+---
+description: Plan fixture-based quality checks and conversion report requirements
+argument-hint: [sample-or-quality-scope]
+allowed-tools: [Read, Glob, Grep, Bash, Edit]
+---
+
+# /plan-quality-evaluation
+
+Plan local fixture evaluation and report requirements for math-heavy PDF conversion.
+
+## Arguments
+
+The user invoked this command with: $ARGUMENTS
+
+## Workflow
+
+1. Read `PLAN.md`, `PROGRESS.md`, `PRD.md`, and `ARCHITECTURE.md`.
+2. Inspect `samples/` only as local fixture context; do not stage or commit sample files.
+3. Define checks for page coverage, reading order, math renderability, delimiter normalization, table handling, asset links, metadata completeness, and warning counts.
+4. Define `.json` metadata and `.report.md` expectations from the same source data.
+5. Separate fast mocked checks from optional MinerU/model/GPU-dependent checks.
+6. Update `PROGRESS.md` with the planned coverage and remaining sample gaps.
+
+## Guardrails
+
+- Do not copy sample PDFs into tracked files.
+- Do not require GPU or large model downloads for the default fast verification loop.

.codex/commands/review-project-docs.md

@@ -0,0 +1,26 @@
+---
+description: Review core project documents for consistency with fixed decisions
+argument-hint: [scope]
+allowed-tools: [Read, Glob, Grep, Bash, Edit]
+---
+
+# /review-project-docs
+
+Review project documents for contradictions, stale decisions, and missing constraints.
+
+## Arguments
+
+The user invoked this command with: $ARGUMENTS
+
+## Workflow
+
+1. Read `PLAN.md` and `PROGRESS.md`.
+2. Read the requested document scope, defaulting to `AGENTS.md`, `PRD.md`, `ARCHITECTURE.md`, and `docs/KNOWLEDGEBASE.md`.
+3. Check for contradictions against fixed decisions: MinerU 3.1.0 only, local-only, direct CLI execution, CLI-internal temporary local `mineru-api` allowed, no `--api-url` or remote API path, Python 3.12, uv, Obsidian Markdown, metadata JSON, and `.report.md`.
+4. Report findings first with file and line references.
+5. If edits are requested, make only surgical documentation changes and update `PROGRESS.md`.
+
+## Guardrails
+
+- Do not add speculative features, alternate engines, web UI, cloud OCR, or manual review queues.
+- Do not rewrite unrelated prose while fixing one inconsistency.

.codex/commands/run-mineru-research.md

@@ -0,0 +1,29 @@
+---
+description: Research current MinerU 3.1.0 facts for local integration planning
+argument-hint: [research-question]
+allowed-tools: [Read, Glob, Grep, Bash, WebFetch, Edit]
+---
+
+# /run-mineru-research
+
+Research MinerU 3.1.0 facts that affect this project's documentation or future implementation.
+
+## Arguments
+
+The user invoked this command with: $ARGUMENTS
+
+## Workflow
+
+1. Read `PLAN.md`, `PROGRESS.md`, `ARCHITECTURE.md`, and `docs/KNOWLEDGEBASE.md`.
+2. Use official MinerU documentation, the MinerU GitHub repository, primary papers, and official dependency documentation.
+3. Verify facts that can change: install commands, supported Python/CUDA versions, CLI flags, output formats, model download behavior, and licenses.
+4. Record sources with URLs and access dates when updating docs.
+5. Keep findings scoped to MinerU 3.1.0; do not add candidate-engine comparisons.
+6. Update `PROGRESS.md` with what was verified and what remains uncertain.
+
+## Guardrails
+
+- Allow only direct `mineru` CLI execution and the CLI-internal temporary local `mineru-api` process.
+- Do not add cloud OCR, hosted LLM, `--api-url`, remote API, router, HTTP client backend, or remote OpenAI-compatible backend paths.
+- Do not turn research notes into implementation code.
+- If official sources conflict, stop and ask for a decision instead of guessing.

.codex/commands/start-agent-work.md

@@ -0,0 +1,32 @@
+---
+description: Start a project task by loading shared plan and progress context
+argument-hint: [agent-or-task]
+allowed-tools: [Read, Glob, Grep, Bash, Edit]
+---
+
+# /start-agent-work
+
+Start work in this repository with the project coordination protocol.
+
+## Arguments
+
+The user invoked this command with: $ARGUMENTS
+
+## Workflow
+
+1. Read `PLAN.md` and `PROGRESS.md`.
+2. State the current goal, the next action, and any blocker that matters for the task.
+3. Read only the additional source documents needed for the requested work.
+4. If subagents are useful and the user explicitly asked for delegated agent work, choose the smallest set of `.codex/agents/*.toml` roles that covers the task.
+5. For substantial implementation work, use the harness sequence: `harness-planner-agent` drafts the plan and contract, `feature-generator-agent` implements one agreed chunk, and `evaluation-agent` reviews the contract and completed work.
+6. Do not implement converter code unless the user explicitly requests implementation.
+7. After meaningful changes, update `PROGRESS.md`; update `PLAN.md` only when sequencing, decisions, ownership, or blockers change.
+8. Run the smallest useful verification, check git status, and commit project changes while excluding `samples/`.
+
+## Guardrails
+
+- Keep MinerU 3.1.0 as the only conversion engine.
+- Allow MinerU 3.1.0's CLI-internal temporary local `mineru-api`, but prohibit `--api-url`, remote APIs, router mode, HTTP client backends, and remote OpenAI-compatible backends.
+- Keep runtime processing local-only.
+- Keep `samples/` out of commits unless the user explicitly requests otherwise.
+- Prefer official sources for changing facts about Codex, MinerU, Python, uv, CUDA, or licenses.