dispatch
Skill2 starsChanged 11 days agoMITView on GitHub
Parent-thread playbook for working with wardens — route work to the owning warden, dispatch with a MissionBrief, act on the UpwardReport with a deterministic state machine, verify, and integrate. Use whenever an edit falls inside a warden's authority, or when a warden returns.
42 other projects use a version of this.
Adopted by
Most recently taken up by
- warrendeanlangeveldt/code-kittheir own version, first seen Nov 15, 2026
- renzrollon/interlocktheir own version, first seen Nov 5, 2026
- dan-myles/skillstheir own version, first seen Nov 1, 2026
- bestony/skillstheir own version, first seen Nov 1, 2026
- AcKeskin/contexturetheir own version, first seen Nov 1, 2026
- mhlaghari/stuntmantheir own version, first seen Nov 1, 2026
Used with
Projects with it also use edge-detection, odds-analysis and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
noahowshSkilledge-detection
Comparing a tennis model's probability with the market: no-vig close, expected value at the best available price, closing line value, fractional Kelly sizing, and the checks that separate an estimated edge from noise. Use when the user has a probability and wants to know whether a price is better than fair, how large a stake is reasonable, or whether an edge is real. Requires a probability the user trusts. Do not use for fetching prices: see odds-explorer. Do not use for devig math alone: see odds-analysis. Do not use for historical testing: see backtesting.
Also in 4 projects that use this
- noahowshSkill
odds-analysis
Odds math for two-way tennis markets: American to implied probability, margin, devigging (multiplicative, power, Shin), no-vig fair odds, consensus across books, favorite-longshot bias, and expected value of a price. Use when the user asks what a price implies, how to remove the vig, what the true probability is, or how to compare books. Do not use for fetching or comparing live prices: see odds-explorer. Do not use for deciding a bet against a model: see edge-detection.
Also in 4 projects that use this
agiprolabsSkillwalk-forward-validation
Walk-forward validation framework for trading strategies and ML models with time-series-aware splits, overfit detection, and regime-aware validation
Also in 4 projects that use this
Yahmez3Skillbacktesting
Use when backtesting a trading strategy — evaluating a signal series against historical price data and computing performance metrics (Sharpe, Sortino, max drawdown, CAGR, win rate, expectancy). Triggers on "backtest this strategy", "what's the Sharpe on X", "run a walk-forward test", or "compare strategies A and B". Produces equity curves, trade logs, and tear sheets.
Also in 4 projects that use this
ffsshhttiikkSkillfeature-engineering
Machine learning feature engineering
Also in 4 projects that use this
varun29ankuSSkillorchestrate
This skill should be used when the user asks to 'orchestrate a task', 'break down work into parallel agents', 'coordinate subtasks', 'run agents in parallel', or mentions 'multi-agent'. Decomposes complex tasks into tracked subtasks, dispatches parallel subagents, and coordinates until completion.
Also in 4 projects that use this
- +3 more from noahowsh
Other versions42 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 42
- your-papa/obsidian-Smart2Brain★ 1.3ksame name
- s2b-dev/smart-second-brain★ 1.3ksame name
- bassimeledath/dispatch★ 411same name
- renzrollon/interlock★ 61same name
- fagemx/edda★ 36same name
- phj128/autoworker★ 18same name
- clonable-eden/cekernel★ 17same name
- mhlaghari/stuntman★ 16same name
- YippeeXuXu/dsh-collab-mode★ 5same name
- BigY0shi/super-skills★ 4same name
32 more not listed
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 repository3 other files it gives its agents.
- mission-briefSkill
- upward-reportSkill
- wardensSkill
The file
We split this skill into the rules it gives an agent
4 rules. 0 are common, 1 is distinctive, and 3 are very specific to this project.
0
Common
Most of the 40 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 23
Write the brief with the mission-brief skill and put it at the top of the Agent prompt.
- Very specific: line 44
After a non-trivial mission, or any surprise, dispatch debrief-scribe with the report and the commit sha.
- Very specific: line 45
Check the team's numbers now and then: warden stats.
- Very specific: line 46
Run warden doctor in CI so stale anchors, overlaps, and expired pilots fail the build instead of sitting unnoticed.
“Common” is measured: kept by at least 6 in 10 of the 40 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
Dispatch — the parent's side of the contract
You're the commander. Wardens own their domains; you own the mission as a whole. That covers routing, verifying, integrating, and escalating.
1. Route
node <plugin>/bin/warden.mjs route <path> [<path>...]- One owner: dispatch that warden.
- Several paths with different owners: split the mission, one brief per warden. Dispatch independent ones in parallel.
- UNOWNED: you do it directly. If the same unowned area keeps coming up, scaffold a warden (
warden new). - Ambiguous (two owners): fix the overlap first (
warden doctorflags it). Ambiguous ownership is how nobody ends up owning anything.
2. Dispatch
Write the brief with the mission-brief skill and put it at the top of the Agent prompt. The dispatch hook denies invalid briefs and records valid ones in the ledger.
3. Act on the UpwardReport (deterministic — no extra LLM review)
| Report | Your action |
completed, confidence ≥ 0.7 | Verify: re-run the brief's success criteria yourself, and check that git diff --name-only is inside the brief's paths. Pass → integrate. |
completed, confidence < 0.7 | Verify as above. If the change adds a mechanism or a default-on behavior, send it to evidence-warden before you integrate. Otherwise accept with the low confidence disclosed. Never re-prompt the warden for self-review. |
failed, retries remain | Re-dispatch the same mission: attempt + 1, the report's blockers copied into prior-blockers, and key-tasks adjusted. |
failed, retries exhausted, or escalation-required | Escalate to the user with the blockers. Don't retry silently. |
denied-by-authority | Run warden route on the blocked path. Owner exists → dispatch that owner with the out-of-scope part. No owner → escalate. Widening a warden's authority is the user's decision. |
blocked | Show the blocker to the user. Don't re-dispatch. |
needs-approval | Put planned-actions in front of the user. If approved, re-dispatch with them under approved-actions. |
| Malformed (hook already sent it back once) | Treat it as failed with confidence 0. Verify everything yourself. |
4. Integrate
Wardens don't commit. You do: one concern per commit, with the warden's evidence-anchors in the message body. If a mission produced two concerns, make two commits.
5. Close the loop
- After a non-trivial mission, or any surprise, dispatch
debrief-scribewith the report and the commit sha. - Check the team's numbers now and then:
warden stats. A warden with a low first-pass rate, frequent guard denials, or manydenied-by-authorityreports has a badly sized authority or a stale primer. Fix the definition, not the symptom. - Run
warden doctorin CI so stale anchors, overlaps, and expired pilots fail the build instead of sitting unnoticed.
Anti-patterns
- ❌ Asking a warden to review its own work. It recreates a verify-retry loop with no new signal.
- ❌ Re-dispatching past
retries-allowed. The hook blocks it; don't work around it with a newtaskslug. - ❌ Widening authority in the brief. The hook blocks it. Edit the warden definition, and only with user sign-off.
- ❌ Adding a new warden without evidence that the team needs it. Grow the roster from
statsandroutemisses, not by intuition. - ❌ Specialists (evidence, release, probe roles) editing source. They surface findings, and domain wardens fix them.
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.