Skip to content
Back to skills

Run Github Project

ASecurity

Use when asked to set up, review, or operate a repository's GitHub Project workflow, including ready claims, role-labelled human work, unknown remote mutation outcomes, Backlog triage, epics, checkpoints, next-issue execution, or an authorized drain.

  • 1,070 stars
  • 0 votes
  • 0 copies
  • 4 views
  • Added September 5, 2026
businessrustgit

Works with

  • terminal

Security analysis

A100/100

Pro scans all 14 files and shows the line behind each finding

Scanned September 30, 2026

npx -y skills add chrisbanes/skills --skill run-github-project --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Run Github Project?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Run Github Project
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/chrisbanes-run-github-project/badge)](https://www.skillsdirectory.com/skills/chrisbanes-run-github-project)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: run-github-project
description: Use when asked to set up, review, or operate a repository's GitHub Project workflow, including ready claims, role-labelled human work, unknown remote mutation outcomes, Backlog triage, epics, checkpoints, next-issue execution, or an authorized drain.
compatibility: "External skill providers are mode-specific: review and setup require none; execution, triage, and Wayfinder lanes use the providers documented in references/workflow-providers.md."
disable-model-invocation: true
---

# Run GitHub Project

The Project is the live control plane. Apply these invariants throughout:

1. **Live authority:** claims, selection, and finish decisions require complete,
   fresh GitHub and Project state; cache and partial reads are hints only.
2. **Controller ownership:** only the controller claims, assigns, mutates shared
   Project state, merges, closes issues, and reconciles. A ticket agent owns its
   worktree, branch, and non-merge PR mutations only.
3. **Unknown outcomes:** reconcile a failed or timed-out remote mutation before
   retrying or reporting success.
4. **Preservation:** retain blocked, dependency-gated, and role-labelled human work in
   its authoritative frontier or partial-drain report; never change state merely
   to make the queue appear empty.

Preserve verified plan state through contract-preserving replans, return true
human work to Backlog, and in `drain` pair occupied slots with warm worktrees
and persistent ticket agents. Park only qualifying terminal required-CI claims
outside capacity before refreshing the control plane.

## Select the mode

- `review`: inspect or explain only; no operation.
- `setup`: configure, validate, or incrementally update the binding only; no
  Project work.
- `next`: default execution; process at most one selected issue.
- `drain`: only on explicit drain/run-all/repeat/until-empty request.

A named Wayfinder child remains `next`; it grants neither drain authority nor a
claim bypass. Before any mode-specific action, read the matching mandatory lane:

| Mode | Mandatory reference |
| --- | --- |
| `review` or `setup` | [Review and setup](references/review-and-setup.md) |
| `next` or `drain` | [Review and setup](references/review-and-setup.md) for binding validation, then [execution controller](references/execution-controller.md) and [ticket lifecycle](references/ticket-lifecycle.md) |

For execution, read [workflow providers](references/workflow-providers.md)
before preconditions; it is authoritative on required, conditional, optional
providers, sources, installation commands, and lane-specific fallback. Never
install a provider implicitly. Read all providers and specialist contracts
required by the execution-controller lane before their relevant action.
When the trusted configuration enables agent profiles, read
[agent routing](references/agent-routing.md) before assigning a planner, ticket
agent, or helper.
For `drain`, read [drain scheduler](references/drain-scheduler.md) completely
before drain queue work. For `next` or `drain`, read
[review contracts](references/review-contracts.md) completely before acceptance
work.

## Cross-mode boundaries

`review` may inspect only permitted repository, supplied, and read-only remote
state. It never configures, ranks, claims, transitions, triages, plans,
delegates, mutates, pushes, merges, or closes. Finish `review-complete` with
evidence, safe next action, and uncertainty, or `review-blocked`.

`setup` reads the configuration required to produce and validate a binding but
requires neither execution dependencies or authority nor clean execution state.
It never ranks/claims, changes Project/issue/PR state, creates worktrees, plans,
implements, pushes, or merges. It finishes only `configuration-valid`,
`configuration-ready-to-commit`, or `configuration-blocked`.

`next` and `drain` require the controller and lifecycle lanes. Standing authority
expires on stop, timeout, crash, or interruption. Do not support publish-only
mode or impose a skill-defined ticket cap in `drain`.

## Final report

The selected lane defines terminal state and report evidence. `setup` reports
only identity, configuration files/read-or-changed, live validation, unresolved
values, committed-base state, and one configuration result. Lead an execution
report with the outcome, completed tickets, work still moving, exact blockers,
and the user's next action, if any. Link to durable issue, PR, and checkpoint
evidence instead of repeating unchanged queries or no-op events. Keep mode,
capacity, configuration digest, live-query and authority evidence, routing
ledger, frontiers, parked/triage/reconciliation outcomes, and per-ticket
selection, lease/plan, branch/commit/PR, verification/review, reconciled
mutation/merge, and preserved or cleaned state available in a concise appendix
when needed to audit a decision or resume work.

Files in this skill

  • SKILL.md38.3 KB
  • agents/openai.yaml491 B
  • references/drain-scheduler.md20.6 KB
  • references/human-frontier.md4.4 KB
  • references/normalized-ticket.md14.7 KB
  • references/planning-lane.md14.3 KB
  • references/project-config.md4.6 KB
  • references/remote-reconciliation.md1.7 KB
  • references/review-contracts.md1.7 KB
  • references/triage-lane.md5.2 KB
  • references/wayfinder-lane.md10.7 KB
  • references/workflow-providers.md2.6 KB
  • scripts/rank_tickets.py67.2 KB
  • scripts/test_rank_tickets.py72.5 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…