raktim-mondol/dsh-researchcraft/ah-revise-from-review
Claude Code and agent skill by raktim-mondol/dsh-researchcraft
Trust tier 3
- community
Source check pending
Checking that the SKILL.md file is still in the repository.
Revise an Academic Harness manuscript in response to reviewer or supervisor comments, then build a response letter in which every claimed change maps to a real edit. Use when asked to address reviews, comments or a revision round.
Install
npx vskill@latest install raktim-mondol/dsh-researchcraft/ah-revise-from-reviewSKILL.md
Revise from review
Academic Harness on DeepSeek Harness. Native ah_* tools. Pi is not required.
- List the items.
ah_reviewshows each numbered item with its status. If the comments are not imported yet, callah_reviewwithaction: import, a reviewid(e.g.R1), andfilepointing at the comments; the harness stores a snapshot of the text at that moment. - One item at a time. For each item decide: change the text (addressed), or explain why not (rebutted). Declining a point is the author's decision: use
ah_decision_request. - Edit only what the item concerns. Find the units with
ah_claims,ah_sweepor by reading. Edit withedit; the harness checks every edit. Numbers stay\fact{...}; claims about a paper followah-write-unit. After changing a number or term, runah_sweepon the old value so no copy is left elsewhere. - Map and respond. Call
ah_review_setwith the item, the unit ids you changed (they are printed in the edit results and inah_checkoutput),status: addressed, and a response in the voice of the authors that says what changed and where. For a rebuttal usestatus: rebuttedand give the reason and, if it applies, the unit where the text already says it. A response that says "we added..." must have a mapped unit whose text really changed, or the harness refuses it. - Build the letter.
ah_review_respondwrites it from the items and the real diffs since the snapshot. Read what it reports: items marked addressed with no change, missing responses, units that changed but belong to no item. Fix those, then build again. - Report to the author: items addressed, items rebutted (with the reason), items waiting on a decision, and anything the letter flagged.
Security
- Security scanPassed, 90 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 repositoryCheckingskills/ah-revise-from-review/SKILL.md
Popularity
- Repository stars
- 1
- 7-day trend score
- 8
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/raktim-mondol/dsh-researchcraft/ah-revise-from-review)