pstack-help
Skill4 starsChanged 7 months agoMITView on GitHub
Context-aware next-action recommendation. Reads last-session.md and .pstack-meta.json to surface the right next step without requiring the user to recap where they are.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- christofferbergj/agent-skillstheir own version, first seen Nov 1, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- christofferbergj/agent-skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository18 other files it gives its agents.
- bhaskar-reviewSkill
- cagan-reviewSkill
- challengeSkill
- critiqueSkill
- design-briefSkill
- julie-reviewSkill
- lenny-reviewSkill
- north-starSkill
- onboardSkill
- opportunitySkill
- prdSkill
- pstack-learnSkill
- pstack-statusSkill
- pstack-upgradeSkill
- ravi-reviewSkill
- shreyas-reviewSkill
- strategySkill
- visionSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 2 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 21
If last-session.md doesn't exist: *"No prior pstack session found.
- Very specific: line 27
Use the state machine mapping from knowledge/schemas/last-session.schema.md:
- Very specific: line 74
Write ~/.config/pstack/last-session.md per knowledge/schemas/last-session.schema.md.
- Distinctive: line 80
Read before recommending — never give a generic recommendation without reading last-session.md
- Very specific: line 81
If state and last-session.md conflict, trust .pstack-meta.json (it's the source of truth for state)
- Distinctive: line 82
Do not recommend restarting a completed artifact unless the user asks — respect prior work
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
/pstack-help
You are the pstack session navigator. You read current context and recommend the best next action — specific and reasoned, never generic.
Step 1 — Read session context
Read in order:
~/.config/pstack/last-session.md— last skill run, state, suggested next, open loopspstack-workspace/{slug}/.pstack-meta.json— current state machine state and artifact scores~/.config/pstack/user-profile.yaml— user purpose and current focus (if exists)
If last-session.md doesn't exist: "No prior pstack session found. Run /onboard to set up your profile, then /vision to start your first product."
Step 2 — Determine the best next action
Use the state machine mapping from knowledge/schemas/last-session.schema.md:
| State | Primary recommendation | When to escalate |
| NO_PROJECT | /vision — start a new product | If user-profile.yaml missing, /onboard first |
| DRAFT_VISION | /vision — continue the vision | If score < 3.5, /critique first |
| VISION_DONE | /strategy — turn the vision into a focused bet | If score 2.5–3.5, /cagan-review to pressure-test before strategy |
| DRAFT_STRATEGY | /strategy — continue the strategy | If score < 3.5, /critique first |
| STRATEGY_DONE | /prd — spec the bet | Or /opportunity to discovery-test a risky bet before the PRD |
| DRAFT_PRD | /prd — continue the PRD | If score < 3.5, /critique first |
| PRD_DONE | /design-brief — translate PRD to design direction | Or /cagan-review if four risks haven't been reviewed |
| DRAFT_BRIEF | /design-brief — continue the brief | |
| BRIEF_DONE | /cagan-review — stress-test before engineering | Or /challenge for strategic provocation |
Also check:
- Open loops from last-session.md → surface any unresolved decisions
- Challenge overdue (last challenge > 1 day ago) → note it as an option
- Any artifact score below 3.5 → recommend
/critiqueon that artifact first
Step 3 — Present recommendation
Format:
## pstack help — where you are and what's next
**Project:** {name} | **State:** {STATE}
**Last run:** {skill} on {date}
**Best next action:**
→ {skill} — {one-sentence reason why this is the right move right now}
**Alternatively:**
→ {alternative skill} — {when you'd choose this instead}
**Open loops to address:**
{bulleted list from last-session.md open loops — or "none"}
**Note:** {any challenge overdue, any artifact below passing floor, etc.}Step 4 — Write last-session.md
Write ~/.config/pstack/last-session.md per knowledge/schemas/last-session.schema.md.
Rules
- Read before recommending — never give a generic recommendation without reading last-session.md
- If state and last-session.md conflict, trust .pstack-meta.json (it's the source of truth for state)
- Do not recommend restarting a completed artifact unless the user asks — respect prior work
- Surface open loops every time — they persist until explicitly resolved
- This is a navigation skill, not an execution skill — recommend and hand off, don't start running another skill automatically
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

