agentleFS
sledtoolssledtools/pika/.agents/skills/land/SKILL.md

land

Skill45 starsChanged 8 months agoMITView on GitHub

Land a branch on the canonical forge using git.pikachat.org and ph. Use when the user asks to upstream, land, merge, submit, or finish a branch; when an agent should push a branch, wait for forge CI, inspect failures, and merge it; or when the user wants the old GitHub-PR landing workflow replaced with the forge-native one.

  • Commits and pushes

25 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use drift-audit, ship and 2 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.
  • mr-fumanchuSkill

    drift-audit

    Audit a project's CLAUDE.md against its memory files to find duplicated or drifting rules, then consolidate them into a single source of truth. Use when Tony says "drift audit", "check for drift", "audit memory vs CLAUDE.md", "find duplicated rules", "clean up memory duplication", or wants to reconcile CLAUDE.md and memory so a rule never lives in two places. Read-only by default; never edits or deletes anything without explicit approval.

    Also in 3 projects that use this

  • AWSSkill

    ship

    Finalize dev work on an issue branch — update docs, push, and open a PR.

    Also in 5 projects that use this

  • SentrySkill

    review

    Fixture review skill.

    Also in 3 projects that use this

  • GoogleSkill

    code-review

    Adversarial systems, concurrency, memory-model, and architecture code review protocol. Use this skill whenever asked to review code, PRs, diffs, or branches in Filament.

    Also in 3 projects that use this

The file

Show its rules and text

We split this skill into the rules it gives an agent

15 rules. 0 are common, 13 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 24 edited versions we read keep these.

  • 13

    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
  • could be automatic: a setting could enforce it
  1. Distinctive:

    Use this skill when the task is to land code on the forge.

    line 8
  2. Very specific:

    wait for forge CI with ph

    line 20
  3. Distinctive:

    Do not default to GitHub PRs.

    line 24
  4. Distinctive:

    Do not silently rename or rewrite a user's remotes.

    line 28
  5. Distinctive:

    Before landing, inspect remotes with:

    line 30
  6. Distinctive:

    If the repo is not set up that way:

    line 41
  7. Distinctive:

    only change remotes if the user explicitly asked

    line 45
  8. Distinctive:

    Do not merge. Fix the issue, push again, and repeat.

    line 68
  9. Distinctive:

    If CI is green:

    line 70
  10. Distinctive:

    Only create a sibling worktree when it materially helps:

    line 89
  11. Distinctive:

    If you do create a sibling worktree:

    line 95
  12. Distinctive:

    keep the current worktree untouched

    line 99
  13. Very specific:

    If no branch or id is passed, ph may infer the current branch.

    line 114
  14. Distinctive:

    If CI appears wedged or the failure is operational rather than code-related, report that clearly instead of pretending the landing path is …

    line 128
  15. Distinctive:

    When finishing a landing task, include:

    line 134

“Common” is measured: kept by at least 6 in 10 of the 24 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Land

Use this skill when the task is to land code on the forge.

This repo is forge-native now:

  • canonical Git is on git.pikachat.org
  • GitHub is a mirror
  • ph is the agent-facing control-plane client

The default path is:

  1. rebase onto canonical master
  2. push branch to the forge
  3. wait for forge CI with ph
  4. inspect failures with ph logs
  5. merge with ph merge

Do not default to GitHub PRs.

Remote policy

Do not silently rename or rewrite a user's remotes.

Before landing, inspect remotes with:

git remote -v

Preferred layout:

  • origin -> git@git.pikachat.org:pika.git
  • github -> git@github.com:sledtools/pika.git

If the repo is not set up that way:

  • tell the user
  • recommend using origin for the forge remote
  • only change remotes if the user explicitly asked

Default landing flow

Assume the current worktree should be used unless the user explicitly wants isolation or continued parallel work.

Typical flow:

git fetch origin
git rebase origin/master
git push -u origin HEAD
ph status
ph wait

If CI fails:

ph logs

Do not merge. Fix the issue, push again, and repeat.

If CI is green:

ph merge

Then verify:

git fetch origin
git rev-parse origin/master
git ls-remote github master

The GitHub mirror check is useful, but do not block on it unless the user explicitly asks or the task is about mirror validation.

When to use a sibling worktree

Only create a sibling worktree when it materially helps:

  • the user wants the current worktree free to keep moving
  • the branch needs careful cherry-pick or rebase cleanup
  • you expect multiple CI fixup pushes and want isolation

If you do create a sibling worktree:

  • make a short random directory in ../
  • branch from origin/master or cherry-pick the local commits missing from origin/master
  • keep the current worktree untouched

ph command guidance

Useful commands:

ph status [branch|id]
ph wait [branch|id]
ph logs [branch|id]
ph merge [branch|id]
ph close [branch|id]
ph url [branch|id]

If no branch or id is passed, ph may infer the current branch. Prefer explicit branch or id when the current worktree state is ambiguous.

Failure handling

Ignore CodeRabbit and Devin unless they point to a concrete independently verified problem.

Focus on:

  • forge CI status
  • required lane failures
  • ph logs
  • real regressions

If CI appears wedged or the failure is operational rather than code-related, report that clearly instead of pretending the landing path is healthy. Do not merge with failing CI unless the user explicitly instructs you to override that guardrail.

Final handoff

When finishing a landing task, include:

  • whether you landed from the current worktree or a sibling worktree
  • which remote you used for the canonical push
  • whether CI passed cleanly or needed fixes
  • whether the branch was merged through ph
  • whether the GitHub mirror caught up if you checked it

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.