agentleFS
youkistudiosyoukistudios/honest-loop/skills/push/SKILL.md

push

Skill2 starsChanged 4 months agoMITView on GitHub

Closes the gaps from the most recent /honest and moves the score toward 9. Fires on "push", "push to 9", "close the gaps", "fix the gaps". Reads /honest's output directly — the user never restates gaps. Executes without faking closure; VALIDATE and DECISION gaps surface, they never get buried.

50 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use get-pr-reviews, push-everything 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.
  • saffron-healthSkill

    get-pr-reviews

    Get nicely formatted output of the last review comments on a PR. Used internally by other skills.

    Also in 3 projects that use this

  • saffron-healthSkill

    push-everything

    Stage ALL changes in the repository (not just session changes), commit, and push. Use when you need to commit all pending changes.

    Also in 3 projects that use this

  • NVIDIASkill

    generate-spec

    Produce a feature specification covering API changes, backward compatibility, cross-language impact, performance, and a file-level change plan. Use when designing a new feature, planning a significant change, or the user asks to spec something out.

    Also in 3 projects that use this

  • ddehartSkill

    knowledge-graph

    Search, capture, and query the claude-code-plugins knowledge graph. Use when asked to "capture this", "add this to the graph", "what do we know about X", "update the graph", "check the graph for X", or when a decision, pattern, or non-obvious fact surfaces mid-session that's worth recording. Also: "log an observation", "what patterns do we have about skill triggering", "any decisions on the knowledge-commons plugin", "what do we know about the commit-creator agent".

    Also in 3 projects that use this

  • anombyte93Skill

    discover

    Phase 1 of the prd-taskmaster pipeline: brainstorm-driven discovery. Delegates to superpowers:brainstorming in Interactive Mode (one adaptive question at a time), or self-brainstorms in Autonomous Mode when no user is present. Intercepts before the brainstorming chain hands off to writing-plans — this skill owns the exit. Extracts constraints, calibrates scale (Solo / Team / Enterprise), and advances the pipeline to GENERATE.

    Also in 4 projects that use this

  • MicrosoftSkill

    merge

    Merge changes from the topic branch to the merge base branch. Use when the user wants to merge their session's work back to the base branch.

    Also in 3 projects that use this

  • +2 more from saffron-health

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 43 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use the most recent /honest output in context (or the named subject's last assessment in ./.honest/log.jsonl if you keep one).

    line 14
  2. Distinctive:

    Never re-run /honest at this stage.

    line 20
  3. Distinctive:

    Before any gap that adds a new tool, service, or file: does a simpler in-stack alternative exist?

    line 32

“Common” is measured: kept by at least 6 in 10 of the 43 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

/push — close the gaps, honestly

/honest evaluates; /push executes. Never the reverse, never merged — the separation is what keeps the scores honest. Vocabulary (gap types, evidence tiers, structural diagnoses) is defined in /honest and not repeated here.

1. Source

Use the most recent /honest output in context (or the named subject's last assessment in ./.honest/log.jsonl if you keep one). Extract BOTH the gap list and the "what makes it a 9" moves — the 9-section holds the leverage, the gap list holds the completeness; close all of it. If the last /honest gave a structural diagnosis (score ≤ 6): stop — push polishes, it cannot re-aim, redesign, re-own, or re-time. Hand the diagnosis back instead. Never re-run /honest at this stage; re-verification comes at the end.

2. Plan (output before executing anything)

One block per gap:

GAP n: [name from /honest]   (TYPE, tested|untested)
File:  [exact path(s) — never "somewhere"]
Move:  [one sentence — what exactly changes]
Blast: [1 file / 2–3 files / 5+ → split, smallest first]

Order: DOC → WIRE → CODE → then surface VALIDATE and DECISION → document DEFERRED. Before any gap that adds a new tool, service, or file: does a simpler in-stack alternative exist? Does something already handle this? Would it touch protected config (hooks, settings, CLAUDE.md)? Protected-config changes surface to the user; never proceed silently.

3. Execute

Per gap: make the change → one-line confirmation → commit if self-contained (fix(push): <gap> — <why>; separate files, separate commits).

Fake-closure prohibition (the load-bearing rule):

  • VALIDATE cannot be closed by writing documentation that says it's validated. Surface: the exact command to run, what success looks like, then continue with the remaining gaps.
  • DECISION cannot be closed by picking the answer yourself. Surface: the exact question and options, then continue.
  • DEFERRED gets its blocker + resolution path written down, then skipped.

Sanity-check gate (before committing any edit to a skill/config that governs behavior): (a) intent match — does the text do exactly what the gap's Move said, no drive-by additions? (b) regression — read 10 lines above and below; does the change contradict existing logic? (c) live path — walk one concrete invocation through the edited section. Any ❌ → fix before staging. This gate exists because a real push once added an exclusion nobody asked for and it shipped unseen.

4. Re-verify

Re-run the full /honest on the subject — the re-verification is the proof, not a summary of it. Same-session re-scores are provisional by definition (a model grading homework it just did); mark them so. The independent confirm is /honest-cycle's verify step — if you're running /push standalone, say plainly: "provisional Y/10 — independent verify pending."

Re-scoreMeaningMove
9+gaps closed, nothing new exposeddone (provisional until independent confirm)
7–8first-order closed, second-order surfacedreport; user decides on another push
≤ 6the gaps were symptomsattach a structural diagnosis; do NOT push again

5. Record the result

Stamp last_pushed, push_score, and provisional: true|false only when the subject is a Markdown file that already begins with YAML frontmatter (an opening --- block). Update that existing block; never prepend frontmatter to source code, JSON, images, generated files, or Markdown without frontmatter.

For every other subject, append the result to ./.honest/log.jsonl using the trajectory format defined in /honest. The log is the safe fallback, not a reason to mutate the artifact being assessed. Nothing else is required; vault-specific trackers are the vault's business.

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.