sischei/deep_learning_for_solving_and_estimating_dynamic_economic_models/strategic-revision
Claude Code and agent skill by sischei/deep_learning_for_solving_and_estimating_dynamic_economic_models
- community
Checking that the SKILL.md file is still in the repository.
Parse one or more referee reports plus the editor letter into a prioritised, parallelisable revision plan with conflict flags and dependency mapping.
Install
npx vskill@latest install sischei/deep_learning_for_solving_and_estimating_dynamic_economic_models/strategic-revisionSKILL.md
/strategic-revision
Description
Turn a batch of referee reports into a prioritised, parallelisable revision plan. Inspired by Jukka Sihvonen's strategic-revision skill.
Given one or more referee reports plus the editor's letter, this skill parses each comment into a discrete task, classifies it, maps dependencies between tasks, flags conflicting requests, and groups everything into parallel execution blocks.
Instructions
You are a meticulous revision manager. When the user invokes /strategic-revision, follow these steps exactly.
Step 1: Collect inputs
- Expect one or more of:
referee1.md,referee2.md,editor.mdin the
working directory. If not present, ask the user to paste the text.
- Also read
paper/main.tex(or the closest manuscript) for section labels.
Step 2: Parse each report into tasks
For every distinct request from a referee, extract:
- Source: which referee / editor comment number
- Verbatim quote (one or two sentences max)
- Task type (tag with exactly one):
argumentative— restructure a claim or add intuitionempirical— new regression, new robustness check, new dataclarification— wording, definitions, framingeditorial— typos, references, formatting- Target location in the paper (section, equation, table, figure)
- Effort estimate (S/M/L)
Step 3: Flag conflicts
Scan all tasks. For each pair, check whether the requests:
- ask for opposite things (e.g. R1 wants more theory, R2 wants less)
- point to the same object with different fixes
Report conflicts explicitly. Do not attempt to silently reconcile them.
Step 4: Map dependencies
Mark task B as blocked by task A when completing A is required to do B (e.g. "rerun regressions" blocks "update table 3").
Step 5: Emit the revision plan
Save to notes/revision_plan.md with this structure:
Step 6: Do not write the response letter yet
This skill produces only the plan. The response letter is a separate step that should be done after the empirical blocks are executed.
Constraints
- Never modify
paper/main.texin this skill; only read it. - Never paraphrase a referee comment beyond the verbatim quote plus classification.
- If a request is ambiguous, mark it
UNCLEARand quote the full sentence. - Keep each task description under 30 words.
Example usage
Expected behavior: parses ~15 comments, produces 3 parallel blocks plus a conflicts section. Execution time: ~2 minutes with Claude Sonnet.
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckinglectures/lecture_06_agentic_programming/code/skills/strategic_revision/SKILL.md
Popularity
- Repository stars
- 172
- Repository forks
- 39
- 7-day trend score
- 26
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/sischei/deep_learning_for_solving_and_estimating_dynamic_economic_models/strategic-revision)