Manage Project Agent Skill

Manage-Project Skill -- Research Project Management

Purpose

Scaffold new research projects, track manuscript writing progress, maintain lightweight project memory, generate pre-submission checklists, and create backwards submission timelines. Integrates with all other writing skills and the project-manager agent.

Commands

/manage-project init {name} --type {type} --journal {journal} [--ssot] [--zotero-collection NAME]

Create a complete project scaffold for a new research paper.

Parameters:

• {name} -- Project identifier (e.g., nnunet-skull-fracture, rfa-meta-analysis)
• --type -- Paper type: original | meta | case | animal | technical | ai_validation | letter
• --journal -- Target journal: RYAI | AJR | Radiology | European_Radiology | KJR | INSI | AJNR | generic
• --ssot -- Emit SSOT.yaml (schema v1) from templates/SSOT.yaml.template instead of legacy project.yaml. Required for Phase 1C auto-enforce (the PostToolUse verify-refs hook blocks instead of warns). New projects should pass --ssot; legacy in-flight projects stay on project.yaml until /manage-project migrate-ssot is run.
• --zotero-collection NAME -- Optional. Create a Zotero collection via pyzotero and populate library_id + collection_key in the contract. Requires ZOTERO_API_KEY + ZOTERO_LIBRARY_ID (optionally ZOTERO_LIBRARY_TYPE, default user). Graceful degrade: with pyzotero missing or credentials absent, the contract is scaffolded with library_id: null / collection_key: null and a WARN is printed.

SSOT template substitutions: {{PROJECT_ID}} → {name}; {{PROJECT_TYPE}} → the SSOT

project_type enum mapped from --type (original → original_research, meta → meta_analysis,

case → case_report, ai_validation → ai_validation, else other). Without

--zotero-collection, library_id / collection_key stay null until the owner links an

existing collection.

Implementation: backed by scripts/init_project.py. Invoke it directly when running outside

the skill harness (from the medsci-skills repo root):

python3 scripts/init_project.py \
    --name {name} --type {type} --journal {journal} [--ssot] \
    --project-root {target_dir}

The helper writes the contract file (SSOT.yaml with --ssot, else legacy project.yaml), the

directory scaffold, the minimal stubs scripts/validate_project_contract.py requires

(manuscript/index.qmd, artifact_manifest.json, qc/status.json), the memory-file templates,

and project_state.json. qc/migration_complete is NOT written by init — that marker belongs

to the migrate pipeline.

Do not hand-build the scaffold. The script is the source of truth for the tree and for

project_state.json; a hand-built one drifts from what the contract validator expects.

Read on demand:

| File | Read it when | Cost if read blindly |

|---|---|---|

| references/init_scaffold.md | you need to know where a scaffolded file lands, or what a project_state.json field means | ~1,400 tokens describing output the script already produces for you |

/manage-project migrate-ssot [--no-mark-complete]

Thin wrapper over scripts/migrate_project_to_ssot.py that converts a legacy project.yaml project into SSOT.yaml form and, by default, touches qc/migration_complete so Phase 1C auto mode switches from warn to enforce.

python3 scripts/migrate_project_to_ssot.py \
    --project-root . --write --mark-complete

• Default: --write --mark-complete (enforce-ready).
• --no-mark-complete flag form: run with --write only. Use when the project still has open QC failures — enforcement is deferred until the migration is validated.
• The migrate script refuses to touch qc/migration_complete unless the generated SSOT.yaml passes validate_project_contract.py AND contract_mode=ssot. Do not touch qc/migration_complete manually.

Re-run after resolving failures; the script is idempotent.

/manage-project status

Report current progress. Reads project_state.json, scans existing files, and checks whether key project memory files are present and aligned.

Output format:

## Project Status: {name}
Journal: {journal} | Type: {type} | Created: {date}
Target submission: {date or "not set"}

### Phase Progress
[check] Phase 0: Project Init (complete)
[check] Phase 1: Outline (complete)
[check] Phase 2: Tables & Figures (complete)
[work]  Phase 3: Methods (IN PROGRESS)
[wait]  Phase 4: Results (pending)
[wait]  Phase 5: Discussion (pending)
[wait]  Phase 6: Introduction & Abstract (pending)
[wait]  Phase 7: Polish (pending)

### Word Counts
Abstract:     0 / 250 words
Introduction: 0 / 600 words
Methods:      847 / 1000 words  85%
Results:      0 / 900 words
Discussion:   0 / 800 words
Total:        847 / 3500 words (journal limit)

### Quality Gates
Critic score (Methods): 87/100 PASS
Citations verified: 12/12
Reporting guideline: pending

### Project Memory
PROJECT.md: present
STATUS.md: stale (last updated 12 days ago)
CLAIMS.md: missing
DATA_DICTIONARY.md: present
ANALYSIS_PLAN.md: missing
REVIEW_LOG.md: n/a (not in revision)

### Next Steps
1. Complete Methods draft (150 words remaining)
2. Run /analyze-stats for Table 1 and diagnostic accuracy
3. Begin Results (Phase 4)

/manage-project sync-memory

Audit and refresh project memory files so other agents can work with less ambiguity.

What it does:

• checks for presence of PROJECT.md, STATUS.md, CLAIMS.md, DATA_DICTIONARY.md, ANALYSIS_PLAN.md, REVIEW_LOG.md
• identifies stale or contradictory project metadata
• proposes the minimum files to create or update
• aligns project_state.json with the current manuscript phase

Suggested use cases:

• before handing the project to project-manager
• before a large revision cycle
• when returning to an old folder after weeks or months
• when multiple collaborators are editing in parallel

/manage-project checklist

Generate a pre-submission checklist covering all quality dimensions.

Output: submission/pre_submission_checklist.md

Read and output the pre-submission checklist from

${CLAUDE_SKILL_DIR}/references/pre_submission_checklist.md.

/manage-project timeline {submission_date}

Generate a backwards timeline from submission date.

Example: /manage-project timeline 2026-05-01

Output:

## Backwards Timeline to Submission: 2026-05-01

Week -8 (2026-03-06): Phase 0-2 complete (scaffold, outline, tables)
Week -7 (2026-03-13): Methods draft -> critic pass
Week -6 (2026-03-20): Results draft -> critic pass + /analyze-stats complete
Week -5 (2026-03-27): Discussion draft -> critic pass
Week -4 (2026-04-03): Introduction + Abstract -> critic pass
Week -3 (2026-04-10): AI pattern removal + /check-reporting (reporting guideline)
Week -2 (2026-04-17): /self-review + co-author review
Week -1 (2026-04-24): Final revisions + figures at 300 DPI + /lit verification
SUBMISSION (2026-05-01): Upload to journal portal

Critical path: Statistics must be complete by Week -6.
Run /analyze-stats as soon as data is available.

Project Scaffold Templates

PROJECT.md Template

# PROJECT

- Title:
- Type:
- Primary question:
- Target journal/venue:
- Lead folder:
- Collaborators:
- Last updated:

STATUS.md Template

# STATUS

- Current stage:
- Current blocker:
- Next actions:
  1.
  2.
  3.
- Last updated:

CLAIMS.md Template

# CLAIMS

| Claim | Supporting result | Source table/figure | Citation status |
|------|-------------------|---------------------|-----------------|
| ...  | ...               | ...                 | ...             |

DATA_DICTIONARY.md Template

# DATA DICTIONARY

| Variable | Definition | Timing | Notes |
|----------|------------|--------|------|
| ...      | ...        | ...    | ...  |

ANALYSIS_PLAN.md Template

# ANALYSIS PLAN

- Primary endpoint:
- Secondary endpoints:
- Main comparator:
- Statistical methods:
- Validation strategy:
- Sensitivity analyses:

REVIEW_LOG.md Template

# REVIEW LOG

| Reviewer comment | Planned action | Status | Location updated |
|------------------|----------------|--------|------------------|
| ...              | ...            | ...    | ...              |

Integration with Other Skills

When called from /manage-project init, automatically:

• Load the correct journal profile from write-paper/references/journal_profiles/{JOURNAL}.md
• Load the correct paper type template from write-paper/references/paper_types/{TYPE}.md
• Copy the appropriate reporting guideline checklist from check-reporting/references/checklists/
• Set word count targets based on the journal profile

After /manage-project checklist, recommend calling:

• /self-review for manuscript quality gate
• /search-lit to verify all citations
• AI pattern removal (built into /write-paper Phase 7)

Project State Conventions

Phase numbers:

• 0 = Init (scaffold created)
• 1 = Outline approved
• 2 = Table/Figure shells approved
• 3 = Methods (critic >= 85)
• 4 = Results (critic >= 85)
• 5 = Discussion (critic >= 85)
• 6 = Introduction + Abstract (critic >= 85)
• 7 = Polish complete (AI pattern removal + checklist + self-review >= 80)

Phase status values: pending | in_progress | complete | blocked

Anti-Hallucination

• Never fabricate file paths, URLs, DOIs, or package names. Verify existence before recommending.
• Never invent journal metadata, impact factors, or submission policies without verification at the journal's website.
• If a tool, package, or resource does not exist or you are unsure, say so explicitly rather than guessing.