Skip to content
Back to skills

Sdd Plan Corrections

ASecurity

SDD corrections when a code review must become traceable tasks; does not implement or alter the report.

  • 3 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added September 19, 2026
ai-agentsc#git

Works with

  • cli

Security analysis

A100/100

Pro scans all 3 files and shows the line behind each finding

Scanned September 29, 2026

npx -y skills add dougcunha/TokenHound --skill sdd-plan-corrections --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Sdd Plan Corrections?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Sdd Plan Corrections
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/dougcunha-sdd-plan-corrections/badge)](https://www.skillsdirectory.com/skills/dougcunha-sdd-plan-corrections)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: sdd-plan-corrections
description: SDD corrections when a code review must become traceable tasks; does not implement or alter the report.
argument-hint: --prd feature-name --num review-number
disable-model-invocation: true
---

# Plan SDD corrections

1. Fix one `tasks/prd-[slug]/codereview_[num]/` by argument or context. When `tasks/prd-[slug]/context-snapshot.md` exists, load it through the Load branch of `.agents/skills/sdd-snapshot/SKILL.md` first. Search only for missing values; multiple candidates require a choice. Require and read its `codereview.md` once; preserve it.
   **Output:** exact, readable report, without mixing IDs between reports.
2. Classify every item as actionable, informational, or pending. In `APPROVED`, plan only what was requested; in `APPROVED WITH RESERVATIONS`, improvements require authorized scope; in `REJECTED`, cover violations, incompleteness, and failures. An unknown status allows only explicitly actionable findings.
   Preserve `CR-NN`; in legacy reports without IDs, assign a local ID by order and section. Inventory root and `done/` metadata to reuse tasks; finding identity is report path + ID.
   **Output:** every item assigned; no duplicate finding or invented decision.
3. Check evidence in the smallest necessary code or TechSpec section; send read-only explorers in parallel only when several findings each need a sweep across many files. Group only the same cause with a reviewable result. Model an acyclic DAG, limits, files, and tests; number new tasks after the largest number in the root and `done/`.
   For desktop C#/.NET, omit E2E; preserve the obligation with relevant unit, integration, or manual acceptance. A missing environment or decision becomes an explicit pending item, not a discarded finding.
   **Output:** every actionable finding has a new or existing task, acceptance, and verification; no orphan task.
4. Read [references/TEMPLATE_TASK.md](references/TEMPLATE_TASK.md) in full when writing. Create reviewable `task_[num].md` files with at least two digits in the report folder, without reusing numbers or overwriting existing tasks. Reference sources by ID and section; preserve handoffs. Do not create a correction manifest: dependencies belong in tasks.
   **Output:** contracts on disk, report intact, and no code changes.
5. Check coverage, traceability, DAG, atomicity, commands, and idempotency. Report files, reuse, pending items, and impact. Return to the caller; request execution approval only where scope is not already authorized. In standalone use, run the session pause from that reference with `sdd-execute-corrections` as next step.
   **Output:** concrete plan for HIL or already authorized execution; contradictions linked to affected tasks.

A task in `done/` whose finding persists is an incomplete correction: keep history and return to the caller to create work in the new report. An unreadable report, conflicting numbering, or cause without evidence blocks only the dependent planning.

Files in this skill

  • SKILL.md2.6 KB
  • agents/openai.yaml43 B
  • references/TEMPLATE_TASK.md2 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.