queue
Skill2 starsChanged 5 months agoMITView on GitHub
Capture a rough future-work idea, interpret it within Engram's current state, split it into one or more scoped tasks in the established `remaining-tasks.md` style, and append them. No commit. Activates when the user types `/queue <thought>`. Does NOT start the work.
7 other projects use a version of this.
Adopted by
Most recently taken up by
- amntoppo/MasterDeck-Terminaltheir own version, first seen Oct 25, 2026
- ARCASSystems/FounderOStheir own version, first seen Oct 24, 2026
- jinhuang712/claude-code-queuetheir own version, first seen Oct 20, 2026
- diogoaraujofx/Claude-Skilltheir own version, first seen Oct 17, 2026
- maxwellsdm1867/wheelertheir own version, first seen Aug 8, 2026
- eduardkumskyi/octotheir own version, first seen Jul 6, 2026
Other versions7 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 7
- maxwellsdm1867/wheeler★ 12same name
- ARCASSystems/FounderOS★ 7same name
- amntoppo/MasterDeck-Terminal★ 4same name
- jinhuang712/claude-code-queue★ 2same name
- breezewish/skills★ 1same name
- eduardkumskyi/octo★ 1same name
- diogoaraujofx/Claude-Skill★ 0same 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 repository5 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- deslopSkill
- dreamSkill
- self-reflectionSkill
- workSkill
The file
We split this skill into the rules it gives an agent
6 rules. 0 are common, 4 are distinctive, and 2 are very specific to this project.
0
Common
Most of the 5 edited versions we read keep these.
4
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 25
Is it architectural?: If it changes storage/sync strategy, the Claude Code integration model, the embedding backend, or module boundaries, …
- Distinctive: line 26
Does it conflict with scope?: No speculative abstractions beyond what the task needs.
- Very specific: line 30
Use AskUserQuestion only if you can't write a concrete acceptance criterion.
- Distinctive: line 44
If the file only has "No tasks queued", replace that note with a fresh section.
- Very specific: line 45
Use Edit (or Write if rebuilding from empty) — never overwrite existing tasks.
- Distinctive: line 55
Start coding or spawn implementer subagents.
“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
/queue — capture future work into remaining-tasks.md
The user has an idea and wants it captured for later. This skill turns it into properly-shaped task entries and appends them to remaining-tasks.md. No implementation happens here.
What you must do
1. Read the project state
remaining-tasks.md— the queue. Match its existing format.CLAUDE.md— operating rules, the ADR requirement, docs-in-sync rule.Sources/,Engram/— skim filenames to understand what's already built.git log --oneline -10— what shipped recently.
2. Interpret the idea
- What does it touch?
EngramCore(store/embeddings/ranking)? theengramCLI? the Xcode app? hooks/skills? docs? - Is it one task or several? One task ≤ ~200 lines net new code and ≤ ~4 files. Split if broader.
- Is it architectural? If it changes storage/sync strategy, the Claude Code integration model, the embedding backend, or module boundaries, it needs an ADR — queue the ADR as the first task in the sequence.
- Does it conflict with scope? No speculative abstractions beyond what the task needs.
3. Ask clarifying questions ONLY when genuinely ambiguous
Use AskUserQuestion only if you can't write a concrete acceptance criterion. Skip if the idea is precise.
4. Draft the task(s)
- **<Imperative title in one sentence>.** <Two-to-four sentences: what's changing, why, which files to read/modify. Be specific.>
- Acceptance: <concrete, testable criteria — "make build && make test pass", "engram <cmd> behaves X", "ADR NNNN written and indexed". Avoid "looks good".>Order tasks by dependency. Put any required ADR first.
5. Append to remaining-tasks.md
- Match the active section if the theme fits; add
## Phase N — <short theme>if meaningfully different. - If the file only has "No tasks queued", replace that note with a fresh section.
- Use
Edit(orWriteif rebuilding from empty) — never overwrite existing tasks.
6. Report and stop
Print: tasks queued (titles), section added to, any clarifications, and "work not started."
Do not spawn an implementer. Do not write code. The skill ends here.
What you must NOT do
- Start coding or spawn implementer subagents.
- Bundle in unrelated edits.
- Reorder or rewrite existing tasks unless the user explicitly asks.
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.