Stats
Actions
Tags
From sws
Sequential orchestrator that runs the SWS review pipeline end-to-end: /sws:verify-claims → /sws:check-fidelity → /sws:peer-review. Passes report paths from the first two into peer-reviewer via explicit CLI args (no autoscan). Propagates SWS_FIDELITY_STATUS to peer-reviewer so the peer-review report transparently notes when fidelity was skipped. Inactive in funding-proposal profile (component skills' profile gates also fire individually).
How this skill is triggered — by the user, by Claude, or both
Slash command
/sws:review-paperThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Run the full SWS review pipeline on the active manuscript.
Run the full SWS review pipeline on the active manuscript.
/sws:review-paper # full pipeline
${PAPER_ROOT}/.sws-project.local.md exists.${CLAUDE_PLUGIN_ROOT}/scripts/agent_prelude.sh review-paper./sws:verify-claims. Wait for completion. Capture exit status. (Skipped silently if profile is funding-proposal.)/sws:check-fidelity. Wait for completion. Read _review/bibliography-fidelity-checker/status.json to determine SWS_FIDELITY_STATUS. (Skipped silently if profile is funding-proposal.)/sws:peer-review --claim-report ${PAPER_ROOT}/_review/claim-verifier/report.md --fidelity-report ${PAPER_ROOT}/_review/bibliography-fidelity-checker/report.md with SWS_FIDELITY_STATUS exported. If the fidelity report file does not exist (e.g. D9c hard skip), omit --fidelity-report and still export SWS_FIDELITY_STATUS=skipped:no-zotero-installation-detected.If RESOLVED_PROFILE_ID == funding-proposal, run peer-reviewer only (claim-verifier and fidelity-checker exit 0 with v0.1-unsupported messages, which is acceptable behavior — orchestrator does not error).
docs/superpowers/specs/2026-05-17-cycle-09-review-design.md — D2, D3, D17.
npx claudepluginhub piripocchio8/scientific-writing-superpowers --plugin swsCreates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.