# Advisor

> |

- **Type:** Skill
- **Install:** `agentstack add skill-upsolve-labs-upstack-advisor`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [Upsolve-Labs](https://agentstack.voostack.com/s/upsolve-labs)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [Upsolve-Labs](https://github.com/Upsolve-Labs)
- **Source:** https://github.com/Upsolve-Labs/upstack/tree/main/skills/advisor
- **Website:** https://upsolve.ai/

## Install

```sh
agentstack add skill-upsolve-labs-upstack-advisor
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

## Update Check (run first)

```bash
_UPD=$(~/.claude/skills/upstack/bin/upstack-update-check 2>/dev/null || .claude/skills/upstack/bin/upstack-update-check 2>/dev/null || true)
[ -n "$_UPD" ] && echo "$_UPD" || true
```

If output shows `UPGRADE_AVAILABLE  `: read `~/.claude/skills/upstack/skills/upgrade/SKILL.md` and follow the "Inline upgrade flow" (auto-upgrade if configured, otherwise AskUserQuestion with 4 options, write snooze state if declined). If `JUST_UPGRADED  `: tell user "Running upstack v{to} (just updated!)" and continue.

# Advisor

You are helping the user decide what to do next.

## Gather Context (silently, do not dump raw output)

1. `git status --short` — uncommitted changes?
2. `git branch --show-current` — which branch?
3. `git log --oneline -5` — recent commits
4. `gh pr list --state open --author @me --limit 5` — open PRs (skip if gh unavailable)
5. Read TODOS.md if it exists
6. If Linear MCP available: check assigned issues

## Categorize State

- **DIRTY**: uncommitted changes -> suggest /execute (to finish + commit) or /validate
- **BEHIND**: branch behind remote -> suggest pull first
- **READY_TO_SHIP**: clean, tests pass, PR approved -> suggest /ship-pr
- **NEEDS_REVIEW**: open PR without review -> suggest /review
- **FRESH**: clean main, no WIP -> suggest /plan
- **IN_PROGRESS**: feature branch with commits -> suggest /validate or /execute

## Present Recommendation

Use AskUserQuestion with:

- 2-line summary of current state
- Primary recommendation with reasoning
- 3-4 alternatives including "plan a new task"

For each option, tell the user exactly what to type (e.g., "Run /plan to start a new feature").

## Rules

- Never run other skills directly. Just recommend.
- Always include /plan as an option.
- If TODOS.md has active items, mention them.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [Upsolve-Labs](https://github.com/Upsolve-Labs)
- **Source:** [Upsolve-Labs/upstack](https://github.com/Upsolve-Labs/upstack)
- **License:** MIT
- **Homepage:** https://upsolve.ai/

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-upsolve-labs-upstack-advisor
- Seller: https://agentstack.voostack.com/s/upsolve-labs
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.