Skip to content
Back to skills

Consult Plan

ASecurity

Grill a proposed change, deviation, or new idea against a repo's existing plan (issues, board, roadmap, decisions), then record the outcome. Use to pitch an idea; update-progress calls it for handoff deviations.

  • 17 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added October 18, 2026
ai-agentsgit

Works with

  • cli

Security analysis

A92/100
  • mediumInstalls packages at runtime which could introduce malicious dependencies

Pro scans all 2 files and shows the line behind each finding

Scanned October 18, 2026

npx -y skills add KrishP147/skills --skill consult-plan --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Consult Plan?

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

Security grade badge for Consult Plan
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/krishp147-consult-plan/badge)](https://www.skillsdirectory.com/skills/krishp147-consult-plan)

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: consult-plan
description: Grill a proposed change, deviation, or new idea against a repo's existing plan (issues, board, roadmap, decisions), then record the outcome. Use to pitch an idea; update-progress calls it for handoff deviations.
argument-hint: "the idea, deviation, or innovation"
---

0. **Target repo** per [BOARD.md](../next/BOARD.md#target-repo): resolve `origin` (this skill needs it on GitHub), `-R <owner>/<repo>` on every `gh` call, and on a fork ask which repo to write to before any write.
1. **Baseline**, whatever exists: open issues, board state, roadmap doc, the decisions doc (found as in step 3) and the grill transcripts (`skilleddocs/grills/`, or `grills/` beside a tracked register) (cite decisions as `D<k>`, or `ADR <n>` in an ADR folder).
2. **Grill**: call the Skill tool with `"grilling"`, baseline as context. The question tree: does the idea fit, conflict with, or extend the plan, and what does the user want to happen?
3. **Record** once settled:
   - Append one row per decision to the repo's decisions doc: the one `AGENTS.md`/`CLAUDE.md` names; else an existing `docs/decisions.md`; else an ADR folder; else `skilleddocs/decisions.md` (same rule as `grill-docs`, which also gives the precedence when several files name one, the folder paths, the header and row format; an ADR folder gets one new ADR per decision, and marks an old ADR it reverses or narrows `Superseded by NNNN` / `Amended by NNNN`). Source: `consult-plan <local date>`, plus the issue or PR link when there is one. Untracked → exclude `skilleddocs/` per the `handoff-auto` skill's `TRACKING.md` (it also covers the `pip install .` hazard).
   - Idea **accepted** or changed: refines existing work → comment on that issue (the decision ids and a one-line outcome; an untracked register, e.g. on a fork, is unreadable to others: quote each decision's text instead of its id); needs its own work → open an issue and put it on the board at Todo ([BOARD.md](../next/BOARD.md)). No repo to write to (fork, both writes declined per BOARD.md's Target repo) → the decision row is the record: add the idea as a roadmap line if one exists, hand the user the issue title and body, and say so.
   - Idea **rejected**: the decision row is the record. Comment only if a related issue exists; never open an issue for a rejected idea.

Files in this skill

  • README.md4.3 KB
  • SKILL.md2.3 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.