Index of every page on this site: llms.txt
All skills
mvanhorn avatar

/printing-press-retro

@9e1a8d4
Run it yourself

Use when the user asks to retro, run a retrospective, file findings, or improve the Printing Press after a printed-CLI run. Trigger phrases: "retro", "retrospective", "what went wrong", "improve the press", "post-mortem", "lessons learned", "what can we improve", "file a retro", "submit findings". Do not use for printed-CLI polish or a single-CLI fix.

Use this Skill: https://skilld.dev/gh/mvanhorn/cli-printing-press/printing-press-retro

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈94 tokens for metadata: the name and description. ≈736 when used: this file. ≈27k more on demand in 13 files.

Description uses 4.7% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈376 characters, or 4.7%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

/printing-press-retro

Result: A scrubbed retro document and, when the issue gate passes, GitHub issues for Printing Press defects only.

Next consumer: A maintainer or agent who will change the Printing Press (generator, scorer, skills, or binary) — not the printed CLI that just shipped.

Done: The manuscript proof exists and is scrubbed; every work unit is file-new, comment-on-existing, or local-only; no unredacted secret or PII remains; the user has the outcome list.

Intent: "Raised the floor" is analysis language, not a filing bar. File only when the Printing Press must change to stop a current, reproducible, generalizing P1 or P2 defect. Manual iteration on one CLI is expected work, not a finding. In user-facing output say "the Printing Press"; name the subsystem when pointing at a fix.

Boundaries

  • Redact every real secret and PII before quoting. Issue bodies and retro docs are public. Use references/secret-scrubbing.md Layer 0. Never quote a leaked value as evidence of the leak.
  • Don't change the machine by default. The burden of proof is on the finding.
  • No P3. File only P1/P2. Anything else is Skip or Drop, not a low-priority issue. The issue gate files P1/P2 Do work units only.
  • Never upload unscrubbed artifacts. Never modify manuscript or library source trees; scrub copies only.

Authority

Invocation authorizes reading manuscripts/library and writing the retro proof plus scratch copies. It does not authorize GitHub filing or public upload until the user confirms Submit. It does not authorize edits to the printed CLI or to Printing Press source.

Steps

Read references/run-resolution.md first. It resolves $API_NAME, $RUN_ID, $RUN_DIR, $CLI_DIR, and $IN_REPO.

Never execute a phase from memory. When you enter a phase, Read its file from phases/ first.

Step File
1 phases/01-gather-evidence.md
2 phases/02-mine-the-session.md
3 phases/03-triage-candidates.md
4 phases/04-classify-findings.md
5 phases/05-prioritize.md
6 phases/06-write-the-retro.md
7 phases/07-plannable-work-units.md
8 phases/08-issue-gate.md
9 phases/09-package-upload-present.md

Follow each file's final Next: pointer. Late procedure lives in the phase file or a role-named reference it loads (references/artifact-packaging.md, references/issue-template.md, references/secret-scrubbing.md). Do not restate those files here.

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

1 warning2mo3 checks · Risk SAFE
  • Gen Agent Trust Hub2mo

    The skill performs automated retrospectives and logs issues to a GitHub repository. It includes data exposure risks due to uploading project artifacts to a public anonymous file-sharing service, along with an indirect prompt injection surface from parsing raw session transcripts.

  • Socket2mo

    1 alert: gptAnomaly

  • Snyk2mo

    Risk: LOW · No issues

Signed by skilld at 9e1a8d4. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 19 hours ago.

Activeupdated 3 months ago
What it can do
Runs commands Reads files Edits files
version
0.1.0
created_by
user
All 7 allowed tools
BashReadGlobGrepWriteAgentAskUserQuestion

README badge

README badge for mvanhorn/cli-printing-press/printing-press-retro