adhoc-plan
Convert user-provided recommendations into a timestamped, execution-ready backlog plan using the active repo profile. Use when a repo needs a one-off plan that should not overwrite PLAN_NEXT.
Best use case
adhoc-plan is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Convert user-provided recommendations into a timestamped, execution-ready backlog plan using the active repo profile. Use when a repo needs a one-off plan that should not overwrite PLAN_NEXT.
Teams using adhoc-plan should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/adhoc-plan/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How adhoc-plan Compares
| Feature / Agent | adhoc-plan | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Convert user-provided recommendations into a timestamped, execution-ready backlog plan using the active repo profile. Use when a repo needs a one-off plan that should not overwrite PLAN_NEXT.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Adhoc Plan This is a local discovery wrapper for the shared Factory skill at `factory/skills/adhoc-plan/SKILL.md`. Before using this skill: 1. Verify `factory/skills/adhoc-plan/SKILL.md` exists. 2. If it is missing, stop and ask the user to run: ```bash git submodule update --init factory ``` Then read `factory/skills/adhoc-plan/SKILL.md` and follow that Factory skill using the active `gait` repo profile unless the user provides another explicit profile. Project wrapper policy: - Write generated plans under profile `plan_output_dir`, currently `product/plans/adhoc`. - Treat the PR as plan-only: run lightweight plan validation only, and do not run full repo validation. - Do not poll, wait for, or inspect PR CI, and do not monitor post-merge CI or run post-merge hotfix loops for plan-only adhoc-plan PRs. - If branch protection, review, permissions, or policy blocks merge, report the blocker instead of waiting on CI. Do not treat this wrapper as the source of truth for plan content. The Factory skill is authoritative for the generated plan structure. Gait repository skill contract: - When command evidence is needed, prefer `gait doctor --json` or another active-profile `gait ... --json` command from `factory/profiles/gait.yaml`. - Require `--json` output for machine-readable command evidence.