[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M781RAVX5M81GDMYSHFYGP1X":3,"skill-same:01M781RAVX5M81GDMYSHFYGP1X":63},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":16,"sourceHtmlUrl":17,"categories":18,"keywords":19,"publishedAt":20,"linkOnly":21,"markdownContent":22,"markdownBody":23,"frontMatter":24,"refName":32,"commitSha":33,"skillDirectory":34,"forksCount":35,"repositoryHtmlUrl":36,"repositoryHomepageUrl":6,"ownerHtmlUrl":37,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":38,"install":39,"resources":52},"01M781RAVX5M81GDMYSHFYGP1X","printing-press-reprint",null,"Regenerate an existing printed CLI from scratch under the current Printing Press, with prior research, prior novel features, and prior patches (post-publish hand-fixes) carried into the writing pipeline as reconciliation context rather than dropped on the floor. Pulls the CLI from the public library if it isn't local, recommends reuse-vs-redo of prior research based on age, then hands off to \u002Fprinting-press with the right context. Use when a machine upgrade would benefit a published CLI more than manual polish. Trigger phrases: \"reprint \u003Capi>\", \"regenerate \u003Capi>\", \"redo the \u003Capi> CLI\", \"rebuild \u003Capi> from scratch\", \"this CLI would benefit from a reprint\".","filippolmt\u002Fskills","filippolmt","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F61194571?v=4","01M7808VK1A8QF8ZXGT2DWTKHS","Filippo Merante Caparrotta","01M7808VKN74HC3EZMK8EV35KS",1,"MIT","https:\u002F\u002Fgithub.com\u002Ffilippolmt\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fprinting-press-reprint\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Ffilippolmt\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fprinting-press-reprint\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: printing-press-reprint\ndescription: >\n  Regenerate an existing printed CLI from scratch under the current Printing\n  Press, with prior research, prior novel features, and prior patches\n  (post-publish hand-fixes) carried into the writing pipeline as\n  reconciliation context rather than dropped on the floor. Pulls the CLI\n  from the public library if it isn't local, recommends reuse-vs-redo of\n  prior research based on age, then hands off to \u002Fprinting-press with the\n  right context. Use when a machine upgrade would benefit a published CLI\n  more than manual polish.\n  Trigger phrases: \"reprint \u003Capi>\", \"regenerate \u003Capi>\", \"redo the \u003Capi> CLI\",\n  \"rebuild \u003Capi> from scratch\", \"this CLI would benefit from a reprint\".\nallowed-tools:\n  - Bash\n  - Read\n  - AskUserQuestion\n  - Skill\ncreated_by: user\n---\n\n# \u002Fprinting-press-reprint\n\nRegenerate an existing printed CLI under the current machine. The user gives\na CLI name and (optionally) reasons for the reprint. This skill ensures the\nprior CLI is locally present, recommends whether to reuse or redo prior\nresearch, and hands off to `\u002Fprinting-press` with the context the\nnovel-features subagent needs to reconcile prior features against the\ncurrent machine — keep, reframe, or drop with reasons, never silent.\n\n```bash\n\u002Fprinting-press-reprint notion\n\u002Fprinting-press-reprint cal.com  the new MCP intent surface landed and the prior CLI ships endpoint-mirror only\n\u002Fprinting-press-reprint allrecipes\n```\n\n## When to run\n\n- A significant Printing Press upgrade (new MCP surface, new auth modes, new\n  transport, scoring rubric changes) would lift this CLI more than manual\n  polish.\n- The published CLI ships with a known systemic gap a reprint would fix.\n- The user wants prior novel features re-evaluated against the current\n  machine and current personas, not carried forward verbatim.\n\nFor one-off code-quality fixes, prefer `\u002Fprinting-press-polish` — it doesn't\nredo research or rebuild the manuscript.\n\n## Setup\n\n```bash\nPRESS_HOME=\"${PRINTING_PRESS_HOME:-$HOME\u002Fprinting-press}\"\nPRESS_LIBRARY=\"$PRESS_HOME\u002Flibrary\"\nPRESS_MANUSCRIPTS=\"$PRESS_HOME\u002Fmanuscripts\"\n\n# Mid-pipeline callers may pass printing_press_bin: \u003Cabs-path> in the args\n# bundle. Prefer it so reprint keeps using the parent skill's preflight-selected\n# binary instead of re-resolving through PATH.\nPRINTING_PRESS_BIN=\"${PRINTING_PRESS_BIN:-}\"\nif [ -z \"$PRINTING_PRESS_BIN\" ] && [ -n \"${ARGUMENTS:-}\" ]; then\n  PRINTING_PRESS_BIN=\"$(printf '%s\\n' \"$ARGUMENTS\" | sed -nE 's\u002F^[[:space:]]*printing_press_bin:[[:space:]]*(.+)$\u002F\\1\u002Fp' | head -1)\"\nfi\nif [ -z \"$PRINTING_PRESS_BIN\" ]; then\n  PRINTING_PRESS_BIN=\"$(command -v cli-printing-press 2>\u002Fdev\u002Fnull || true)\"\nfi\n\nif [ -z \"$PRINTING_PRESS_BIN\" ]; then\n  echo \"cli-printing-press binary not found.\"\n  echo \"Install with:  go install github.com\u002Fmvanhorn\u002Fcli-printing-press\u002Fv4\u002Fcmd\u002Fcli-printing-press@latest\"\n  return 1 2>\u002Fdev\u002Fnull || exit 1\nfi\necho \"PRINTING_PRESS_BIN=$PRINTING_PRESS_BIN\"\n```\n\n## Phase A — Resolve and reconcile presence\n\nResolve the user's argument the same way `\u002Fprinting-press-import` does:\nfetch the public library `registry.json` once, then exact → normalized →\nfuzzy match. The argument can be an API slug (`notion`), a brand name\n(`cal.com`), an old `\u003Capi>-pp-cli` form, or close enough.\n\nCapture from the matched registry entry: `API_SLUG` (from `.name`) and\n`LIB_PATH` (from `.path`, e.g., `library\u002Fproductivity\u002Fcal-com`). Phase B\nuses `$LIB_PATH` for the public patches fetch. For the \"present | absent\"\nnever-published row, `$LIB_PATH` stays empty — Phase B's fetch\nshort-circuits on that.\n\nThen check what exists locally and reconcile against the public library by\nreading both provenance manifests' `run_id` and `generated_at`:\n\n| Local | Public registry | Action |\n|-------|-----------------|--------|\n| absent | absent | STOP — nothing to reprint; suggest `\u002Fprinting-press \u003Capi>` for a fresh print |\n| absent | present | invoke `\u002Fprinting-press-import \u003Capi>`, then continue |\n| present | absent | continue — never-published local CLI; skip import |\n| present, same `run_id` | present | continue without import |\n| present, public newer `generated_at` | present | offer import via `AskUserQuestion`; user decides |\n| present, local newer `generated_at` | present | STOP — local has unpublished work; tell user to publish or discard first |\n\nWhen invoking `\u002Fprinting-press-import`, let it own backup, overwrite,\nbuild-verify, and module-path-rewrite. Wait for it to return clean before\ncontinuing.\n\n## Phase B — Verify reconcilable prior context\n\nLocate the two artifacts the writing pipeline should be aware of: research\n(drives novel-features Pass 2(d)) and patches (post-publish hand-fixes\nrecorded by `\u002Fprinting-press-amend`, e.g. live-discovered API quirks the\nspec didn't reveal).\n\n```bash\nLIB_TARGET=\"$PRESS_LIBRARY\u002F$API_SLUG\"\nLIB_RESEARCH=\"$LIB_TARGET\u002Fresearch.json\"\nMAN_RESEARCH=$(ls -1t \"$PRESS_MANUSCRIPTS\u002F$API_SLUG\"\u002F*\u002Fresearch.json 2>\u002Fdev\u002Fnull | head -1)\n```\n\n### Research absent\n\nIf neither research path exists, the published CLI predates `research.json`\nprovenance. The subagent will treat the run as a first print and Pass 2(d)\nreprint reconciliation will not fire — there is nothing for it to read.\nSurface this and ask:\n\n> Published `\u003Capi>` was built before `research.json` provenance landed.\n> Without it, the novel-features subagent will treat this as a first\n> print — there is nothing to reconcile against. Continue as a degraded\n> reprint (essentially a fresh print with a kept binary name)?\n\nIf the user declines, exit. If they continue, record the absence so the\nhand-off prompt notes that this is a degraded reprint.\n\n### Patches discovery\n\nRefresh the local patches index from public when reachable, then read\nlocally so downstream references are durable. Amends may have landed\nagainst the public copy without triggering a regen, so the local copy\ncan lag even when `run_id` matches; this step closes that gap.\n\nThe index ships in one of two shapes: the per-patch directory\n`.printing-press-patches\u002F` (current) or the legacy single-array\n`.printing-press-patches.json` (older CLIs not yet normalized). Always check\nboth shapes when reachable. Prefer the directory when it contains patch files,\nbut never let an absent legacy single-file index set `PATCH_COUNT=0` until the\ndirectory fallback has been read.\n\n```bash\nPATCHES_DIR=\"$LIB_TARGET\u002F.printing-press-patches\"\nPATCHES_LEGACY=\"$LIB_TARGET\u002F.printing-press-patches.json\"\nif [[ -n \"$LIB_PATH\" ]]; then\n  # Fetch the legacy shape if present, but do not treat a 404 as proof that no\n  # patches exist. Current library entries may carry only the per-patch\n  # directory below.\n  tmp=$(mktemp)\n  if gh api -H \"Accept: application\u002Fvnd.github.v3.raw\" \\\n       \"repos\u002Fmvanhorn\u002Fprinting-press-library\u002Fcontents\u002F$LIB_PATH\u002F.printing-press-patches.json\" \\\n       > \"$tmp\" 2>\u002Fdev\u002Fnull; then\n    mv \"$tmp\" \"$PATCHES_LEGACY\"\n  else\n    rm -f \"$tmp\"\n  fi\n\n  # Fetch the current directory shape independently. This is the required\n  # fallback when `.printing-press-patches.json` is absent.\n  listing=$(gh api \"repos\u002Fmvanhorn\u002Fprinting-press-library\u002Fcontents\u002F$LIB_PATH\u002F.printing-press-patches\" 2>\u002Fdev\u002Fnull || true)\n  if jq -e 'type == \"array\"' \u003C\u003C\u003C\"$listing\" >\u002Fdev\u002Fnull 2>&1; then\n    mkdir -p \"$PATCHES_DIR\"\n    jq -r '.[] | select(.name | endswith(\".json\")) | \"\\(.name)\\t\\(.download_url)\"' \u003C\u003C\u003C\"$listing\" \\\n    | while IFS=$'\\t' read -r name url; do\n        tmp=$(mktemp)\n        if curl -fsSL \"$url\" -o \"$tmp\" 2>\u002Fdev\u002Fnull; then\n          mv \"$tmp\" \"$PATCHES_DIR\u002F$name\"   # atomic: a dropped transfer never leaves corrupt JSON\n        else\n          rm -f \"$tmp\"\n        fi\n      done\n  fi\nfi\n\n# Count from the first non-empty shape locally; PATCHES_SOURCE is what Phase D reads.\nif [[ -d \"$PATCHES_DIR\" ]]; then\n  DIR_PATCH_COUNT=$(find \"$PATCHES_DIR\" -maxdepth 1 -name '*.json' ! -name '_meta.json' | wc -l | tr -d ' ')\nelse\n  DIR_PATCH_COUNT=0\nfi\nif [[ \"$DIR_PATCH_COUNT\" != \"0\" ]]; then\n  PATCH_COUNT=\"$DIR_PATCH_COUNT\"\n  PATCHES_SOURCE=\"$PATCHES_DIR\"\nelif [[ -f \"$PATCHES_LEGACY\" ]]; then\n  PATCH_COUNT=$(jq '(.patches \u002F\u002F []) | length' \"$PATCHES_LEGACY\" 2>\u002Fdev\u002Fnull || echo 0)\n  PATCHES_SOURCE=\"$PATCHES_LEGACY\"\nelse\n  PATCH_COUNT=0\n  PATCHES_SOURCE=\"$PATCHES_DIR\"\nfi\n```\n\nIf `$PATCH_COUNT == 0` or no index is present (older CLI predating the\npatches contract), skip the rest of this subsection — no patches block in\nthe hand-off.\n\nIf `$PATCH_COUNT > 0`, surface a one-liner to the user before continuing:\n\n> Public `\u003Capi>` has `$PATCH_COUNT` recorded patch(es) against the prior\n> printed CLI. Carry them into the brief as a watch-list. Regen and\n> publish-validate now read the records and fail closed if a recorded\n> file or declared call site \u002F `pp:patch` marker is gone — do not treat\n> the index as a claim that the customization still shipped.\n\nHold `$PATCHES_SOURCE` and `$PATCH_COUNT` for Phase D.\n\n## Phase C — Recency recommendation\n\nPull `researched_at` from the most-recent prior `research.json` and\n`printing_press_version` + `generated_at` from `.printing-press.json`:\n\n```bash\nRESEARCHED_AT=$(jq -r '.researched_at \u002F\u002F empty' \"$MAN_RESEARCH\" 2>\u002Fdev\u002Fnull)\nPRESS_VERSION=$(jq -r '.printing_press_version \u002F\u002F empty' \"$LIB_TARGET\u002F.printing-press.json\" 2>\u002Fdev\u002Fnull)\nGENERATED_AT=$(jq -r '.generated_at \u002F\u002F empty' \"$LIB_TARGET\u002F.printing-press.json\" 2>\u002Fdev\u002Fnull)\n```\n\nCompute the calendar age of the research with `python3` so it stays portable\nacross macOS\u002FLinux and tolerates the microsecond precision that\n`generated_at` carries (BSD `date -f` rejects fractional seconds; `python3`\nis on every supported platform):\n\n```bash\nAGE_DAYS=$(python3 -c \"\nfrom datetime import datetime, timezone\nts = '$RESEARCHED_AT'.replace('Z', '+00:00')\nprint(int((datetime.now(timezone.utc) - datetime.fromisoformat(ts)).total_seconds() \u002F\u002F 86400))\n\" 2>\u002Fdev\u002Fnull)\n```\n\nSurface both signals to the user — research age and prior machine version.\nAge thresholds are rules of thumb, not gates:\n\n- under 30 days → reuse looks safe\n- 30–120 days → reuse plausible; the user should mention any known API\n  churn in their reprint reason so the subagent's Pass 2 picks it up\n- over 120 days → redo recommended\n\nDon't predict API churn from age alone — describe the signals and let the\nuser override. The Phase 0 binary-version-bump revalidation in\n`\u002Fprinting-press` handles the machine-delta side independently; don't\nduplicate it here.\n\nAsk via `AskUserQuestion`:\n\n1. **Reuse prior research** — keep the prior brief; the subagent re-scores\n   prior novel features against current personas\n2. **Redo research** — re-run Phase 1 of `\u002Fprinting-press` from scratch; the subagent still\n   ingests prior novel features as Pass 2(d) input\n3. **Show me first** — display the prior brief's headline + novel-features\n   list, then re-ask between options 1 and 2\n\n## Phase D — Hand off to `\u002Fprinting-press`\n\nBefore invoking `\u002Fprinting-press`, use the prior CLI's scorecard and manifest\nto decide whether the reprint should offer spec enrichment that the first print\ncould not have used. Reprints have better evidence than fresh prints: they know\nwhich structural dimensions were weak, which Printing Press version produced\nthe prior CLI, and what the user named as the reason for regenerating.\n\nFind the most recent scorecard JSON from the prior manuscript run. If no\nscorecard artifact exists, run a fresh structural scorecard against the local\nlibrary copy:\n\n```bash\nSCORECARD_SOURCE=$(ls -1t \"$PRESS_MANUSCRIPTS\u002F$API_SLUG\"\u002F*\u002Fproofs\u002Fscorecard.json 2>\u002Fdev\u002Fnull | head -1)\nSCORECARD_JSON=\"\"\nif [[ -n \"$SCORECARD_SOURCE\" ]]; then\n  SCORECARD_JSON=$(cat \"$SCORECARD_SOURCE\" 2>\u002Fdev\u002Fnull || true)\nelif [[ -d \"$LIB_TARGET\" ]]; then\n  SCORECARD_SOURCE=$(mktemp)\n  if \"$PRINTING_PRESS_BIN\" scorecard --dir \"$LIB_TARGET\" --json > \"$SCORECARD_SOURCE\" 2>\u002Fdev\u002Fnull; then\n    SCORECARD_JSON=$(cat \"$SCORECARD_SOURCE\" 2>\u002Fdev\u002Fnull || true)\n  fi\n  rm -f \"$SCORECARD_SOURCE\"\n  SCORECARD_SOURCE=\"\"\nfi\n```\n\nIf `SCORECARD_JSON` is empty, continue without enrichment prompting and say the\nreprint is proceeding without prior score evidence. Do not invent a prompt from\nthe reprint reason alone.\n\nWhen `SCORECARD_JSON` is available, inspect only dimensions that map to a\npre-generation spec edit and skip dimensions that already score 10\u002F10:\n\n- `mcp_remote_transport`, `mcp_token_efficiency`, `mcp_tool_design`, and\n  `mcp_surface_strategy` below 10 can be lifted by the `\u002Fprinting-press`\n  Phase 2 section **Pre-Generation MCP Enrichment**. Examples:\n  - remote transport is below 10: offer `mcp.transport: [stdio, http]` or the\n    OpenAPI `x-mcp.transport` equivalent before regeneration.\n  - token efficiency, tool design, or surface strategy is below 10: offer the\n    `\u002Fprinting-press` Phase 2 MCP surface decision, including intents for clear multi-step\n    workflows or the Cloudflare pattern for large surfaces.\n- `auth_protocol` below 10, or prior manifest evidence that the CLI used a\n  slug-derived env var where the ecosystem has a canonical env var, can be\n  lifted by **Pre-Generation Auth Enrichment**. Offer to carry canonical\n  `auth.env_vars` or OpenAPI `x-auth-env-vars` guidance into the spec.\n- `data_pipeline_integrity` below 10 is only an enrichment opportunity when the\n  prior CLI or research shows sync-eligible resources. In that case, point the\n  handoff at the relevant `\u002Fprinting-press` Phase 2 sync\u002Fcache enrichment decision rather than\n  treating the score alone as proof that a local store should exist.\n\nUse `AskUserQuestion` for each concrete opportunity before the handoff. Phrase\nthe question around the scorecard evidence and the named canonical section, not\naround a freeform rewrite. Example:\n\n> Prior scorecard shows `mcp_remote_transport: 5\u002F10`. Offer MCP transport\n> enrichment before regenerating, using `\u002Fprinting-press` Phase 2\n> **Pre-Generation MCP Enrichment** as the source of truth?\n\nOptions:\n\n1. **Apply enrichment to the handoff** - include the selected spec edit in the\n   `\u002Fprinting-press` prompt so Phase 2 can update the spec before generation.\n2. **Skip for this reprint** - leave the spec unchanged for this dimension.\n3. **Show score evidence first** - print the relevant scorecard lines and then\n   re-ask between options 1 and 2.\n\nDo not auto-apply enrichment. If the prior scorecard already has\n`mcp_remote_transport: 10\u002F10`, do not ask the redundant MCP transport question.\nIf the user accepts any opportunity, add a `## Reprint Spec Enrichment`\nsection to the `\u002Fprinting-press` handoff. Keep it brief: name the weak\ndimension, the accepted enrichment, and the canonical `\u002Fprinting-press` Phase 2\nsection to execute. Do not duplicate the canonical enrichment text here.\n\nInvoke `\u002Fprinting-press \u003Capi>` and bundle these into the prompt:\n\n1. **A header line** stating the user already chose to regenerate, so\n   `\u002Fprinting-press` Phase 0's library-check should select \"Generate a fresh CLI\" and not\n   re-prompt fresh-vs-improve.\n2. **Research mode** from Phase C (`reuse` or `redo`). `\u002Fprinting-press` Phase 0's existing\n   reuse logic consumes this.\n3. **The user's freeform reprint reason**, verbatim, in a `User context`\n   block. This propagates into the brief as `## User Vision` and becomes\n   Pass 2(e) input to the novel-features subagent — the right hook for\n   \"I want better MCP support\" → bias the brainstorm accordingly.\n4. **Reprint spec enrichment** — only when the scorecard-driven prompt above\n   found an accepted opportunity. Include under a\n   `## Reprint Spec Enrichment` heading.\n5. **Prior patches** — only when Phase B found `$PATCH_COUNT > 0`.\n   Include under a `## Prior Patches` heading.\n\n   Lead the section with this framing sentence, verbatim:\n\n   > The following were hand-fixed against the prior printed CLI. They\n   > are informational — the machine may have absorbed some of these\n   > upstream since the patch was applied. Stay aware so the fresh code\n   > doesn't silently regress the underlying API truth or architectural\n   > pattern, but do not treat as a re-apply checklist.\n\n   Then summarize the patches from `$PATCHES_SOURCE`. Lead each entry\n   with the *substance* the patch encodes (the API truth, the\n   architectural pattern, the cross-file convention) drawn from each\n   patch's `reason` and `validated_outcome` — not just `summary`. Patch\n   metadata (`id`, `files`) is fine as parenthetical context but never\n   the headline.\n\n   Scale the section to patch count:\n\n   - **1–3 patches**: one short paragraph per patch — substance first,\n     then how it manifested. Model the shape on:\n     > Linear's personal API keys go in `Authorization: lin_api_…` raw,\n     > with no `Bearer` prefix; OAuth tokens use `Authorization: Bearer\n     > \u003Ctoken>`. The prior CLI shipped with `auth set-token` only\n     > (Bearer-defaulted) and was patched post-publish to add\n     > `auth set-api-key` plus the raw-header path in `config.go`\n     > (`linear-auth-api-key-vs-oauth-token`).\n   - **4–9 patches**: one tight sentence per patch, same substance-first\n     shape.\n   - **10+ patches**: thematic summary by category (auth, pagination,\n     query encoding, MCP additions, helpers, classifiers, error\n     envelopes). Two or three sentences per theme citing 2–3 patch `id`s\n     as evidence. Do not enumerate all entries inline.\n\n   Close the section with a pointer so a downstream agent can drill into\n   any specific patch when working in the relevant area:\n\n   > Full patch detail: the per-patch files under `$PATCHES_SOURCE`\n   > (each `\u003Cid>.json` is one self-contained patch; the legacy\n   > single-array `.printing-press-patches.json` carries the same fields\n   > under `patches[]`).\n\nDo **not** pass a separate \"this is a reprint\" marker. The novel-features\nsubagent runs unconditionally on every print and discovers prior research\nvia its own discovery snippet. The paths\nimport populated in Phase A are exactly the paths it checks; Pass 2(d)\nfires whenever prior `research.json` exists.\n\nThe library-preservation contract is owned by `\u002Fprinting-press` Phase 5.6\n(\"Promote to Library\"), not by this skill. When the existing library has\n`novel_features > 0` in its manifest (or hand-authored files under\n`internal\u002Fcli\u002F`, `internal\u002Fsyncer\u002F`, or `internal\u002Fstore\u002F`), Phase 5.6 first\ndry-runs `\"$PRINTING_PRESS_BIN\" regen-merge \"$LIB_TARGET\" --fresh \"$CLI_WORK_DIR\"\n--json` to decide whether the fresh tree rebuilt the prior novels. If the\nfresh tree contains all prior novel work, Phase 5.6 uses the swap path and\ntreats generated-file version drift as expected overwrite. Otherwise it routes\npromotion through `regen-merge --apply` so still-unique hand-authored novels\nsurvive the reprint and genuine `NOVEL-COLLISION` \u002F missing-referent cases halt\nfor review. This honors the prefer-`regen-merge` guidance under the\n**Hand-edits must be regen-mergeable.** section of\n`\u002Fprinting-press` Phase 3. If a future\nedit to that phase changes the routing rule, update this paragraph in the same\nPR -- the reprint skill is the dominant entry point that fires it.\n\nBefore the hand-off, compare regenerated manifest files against the tracked\npublished tree. If `$LIB_TARGET\u002Fmanifest.json` or\n`$LIB_TARGET\u002Ftools-manifest.json` exists, diff each file against the freshly\ngenerated counterpart under `$CLI_WORK_DIR` and surface non-empty diffs in the\nhandoff prompt. Treat a diff as a reconciliation checkpoint, not as automatic\noverwrite approval: the operator must decide whether to preserve the tracked\nhand-edit, fold it into the spec\u002Fresearch inputs, or intentionally accept the\nregenerated value. Do not continue silently when tracked manifest fields,\ntool metadata, or descriptions would be dropped.\n\nAttribution also stays owned by `\u002Fprinting-press`: the hand-off runs generation\nfor the same API slug, and the generate\u002Fpromote path must preserve the existing\nlibrary manifest's permanent `creator` while adding the reprinter to\n`contributors[]` when they differ. Do not repair this by hand-editing\n`creator`, `contributors[]`, README bylines, SKILL `author:`, or NOTICE; rerun\nwith a current Printing Press binary so the manifest-first guard rewrites the\nworking tree and promoted tree consistently.\n\n## After hand-off\n\nThe printing-press flow drives the rest. Don't summarize its work — let\nthe user see the live phases.\n\nIf `\u002Fprinting-press` halts with the subagent's pre-flight HALT (brief\nlacks concrete `Users` \u002F `Top Workflows` content), the reused prior brief\npredates the subagent's required schema. Recommend re-running with **Redo\nresearch** selected at Phase C.\n","# \u002Fprinting-press-reprint\n\nRegenerate an existing printed CLI under the current machine. The user gives\na CLI name and (optionally) reasons for the reprint. This skill ensures the\nprior CLI is locally present, recommends whether to reuse or redo prior\nresearch, and hands off to `\u002Fprinting-press` with the context the\nnovel-features subagent needs to reconcile prior features against the\ncurrent machine — keep, reframe, or drop with reasons, never silent.\n\n```bash\n\u002Fprinting-press-reprint notion\n\u002Fprinting-press-reprint cal.com  the new MCP intent surface landed and the prior CLI ships endpoint-mirror only\n\u002Fprinting-press-reprint allrecipes\n```\n\n## When to run\n\n- A significant Printing Press upgrade (new MCP surface, new auth modes, new\n  transport, scoring rubric changes) would lift this CLI more than manual\n  polish.\n- The published CLI ships with a known systemic gap a reprint would fix.\n- The user wants prior novel features re-evaluated against the current\n  machine and current personas, not carried forward verbatim.\n\nFor one-off code-quality fixes, prefer `\u002Fprinting-press-polish` — it doesn't\nredo research or rebuild the manuscript.\n\n## Setup\n\n```bash\nPRESS_HOME=\"${PRINTING_PRESS_HOME:-$HOME\u002Fprinting-press}\"\nPRESS_LIBRARY=\"$PRESS_HOME\u002Flibrary\"\nPRESS_MANUSCRIPTS=\"$PRESS_HOME\u002Fmanuscripts\"\n\n# Mid-pipeline callers may pass printing_press_bin: \u003Cabs-path> in the args\n# bundle. Prefer it so reprint keeps using the parent skill's preflight-selected\n# binary instead of re-resolving through PATH.\nPRINTING_PRESS_BIN=\"${PRINTING_PRESS_BIN:-}\"\nif [ -z \"$PRINTING_PRESS_BIN\" ] && [ -n \"${ARGUMENTS:-}\" ]; then\n  PRINTING_PRESS_BIN=\"$(printf '%s\\n' \"$ARGUMENTS\" | sed -nE 's\u002F^[[:space:]]*printing_press_bin:[[:space:]]*(.+)$\u002F\\1\u002Fp' | head -1)\"\nfi\nif [ -z \"$PRINTING_PRESS_BIN\" ]; then\n  PRINTING_PRESS_BIN=\"$(command -v cli-printing-press 2>\u002Fdev\u002Fnull || true)\"\nfi\n\nif [ -z \"$PRINTING_PRESS_BIN\" ]; then\n  echo \"cli-printing-press binary not found.\"\n  echo \"Install with:  go install github.com\u002Fmvanhorn\u002Fcli-printing-press\u002Fv4\u002Fcmd\u002Fcli-printing-press@latest\"\n  return 1 2>\u002Fdev\u002Fnull || exit 1\nfi\necho \"PRINTING_PRESS_BIN=$PRINTING_PRESS_BIN\"\n```\n\n## Phase A — Resolve and reconcile presence\n\nResolve the user's argument the same way `\u002Fprinting-press-import` does:\nfetch the public library `registry.json` once, then exact → normalized →\nfuzzy match. The argument can be an API slug (`notion`), a brand name\n(`cal.com`), an old `\u003Capi>-pp-cli` form, or close enough.\n\nCapture from the matched registry entry: `API_SLUG` (from `.name`) and\n`LIB_PATH` (from `.path`, e.g., `library\u002Fproductivity\u002Fcal-com`). Phase B\nuses `$LIB_PATH` for the public patches fetch. For the \"present | absent\"\nnever-published row, `$LIB_PATH` stays empty — Phase B's fetch\nshort-circuits on that.\n\nThen check what exists locally and reconcile against the public library by\nreading both provenance manifests' `run_id` and `generated_at`:\n\n| Local | Public registry | Action |\n|-------|-----------------|--------|\n| absent | absent | STOP — nothing to reprint; suggest `\u002Fprinting-press \u003Capi>` for a fresh print |\n| absent | present | invoke `\u002Fprinting-press-import \u003Capi>`, then continue |\n| present | absent | continue — never-published local CLI; skip import |\n| present, same `run_id` | present | continue without import |\n| present, public newer `generated_at` | present | offer import via `AskUserQuestion`; user decides |\n| present, local newer `generated_at` | present | STOP — local has unpublished work; tell user to publish or discard first |\n\nWhen invoking `\u002Fprinting-press-import`, let it own backup, overwrite,\nbuild-verify, and module-path-rewrite. Wait for it to return clean before\ncontinuing.\n\n## Phase B — Verify reconcilable prior context\n\nLocate the two artifacts the writing pipeline should be aware of: research\n(drives novel-features Pass 2(d)) and patches (post-publish hand-fixes\nrecorded by `\u002Fprinting-press-amend`, e.g. live-discovered API quirks the\nspec didn't reveal).\n\n```bash\nLIB_TARGET=\"$PRESS_LIBRARY\u002F$API_SLUG\"\nLIB_RESEARCH=\"$LIB_TARGET\u002Fresearch.json\"\nMAN_RESEARCH=$(ls -1t \"$PRESS_MANUSCRIPTS\u002F$API_SLUG\"\u002F*\u002Fresearch.json 2>\u002Fdev\u002Fnull | head -1)\n```\n\n### Research absent\n\nIf neither research path exists, the published CLI predates `research.json`\nprovenance. The subagent will treat the run as a first print and Pass 2(d)\nreprint reconciliation will not fire — there is nothing for it to read.\nSurface this and ask:\n\n> Published `\u003Capi>` was built before `research.json` provenance landed.\n> Without it, the novel-features subagent will treat this as a first\n> print — there is nothing to reconcile against. Continue as a degraded\n> reprint (essentially a fresh print with a kept binary name)?\n\nIf the user declines, exit. If they continue, record the absence so the\nhand-off prompt notes that this is a degraded reprint.\n\n### Patches discovery\n\nRefresh the local patches index from public when reachable, then read\nlocally so downstream references are durable. Amends may have landed\nagainst the public copy without triggering a regen, so the local copy\ncan lag even when `run_id` matches; this step closes that gap.\n\nThe index ships in one of two shapes: the per-patch directory\n`.printing-press-patches\u002F` (current) or the legacy single-array\n`.printing-press-patches.json` (older CLIs not yet normalized). Always check\nboth shapes when reachable. Prefer the directory when it contains patch files,\nbut never let an absent legacy single-file index set `PATCH_COUNT=0` until the\ndirectory fallback has been read.\n\n```bash\nPATCHES_DIR=\"$LIB_TARGET\u002F.printing-press-patches\"\nPATCHES_LEGACY=\"$LIB_TARGET\u002F.printing-press-patches.json\"\nif [[ -n \"$LIB_PATH\" ]]; then\n  # Fetch the legacy shape if present, but do not treat a 404 as proof that no\n  # patches exist. Current library entries may carry only the per-patch\n  # directory below.\n  tmp=$(mktemp)\n  if gh api -H \"Accept: application\u002Fvnd.github.v3.raw\" \\\n       \"repos\u002Fmvanhorn\u002Fprinting-press-library\u002Fcontents\u002F$LIB_PATH\u002F.printing-press-patches.json\" \\\n       > \"$tmp\" 2>\u002Fdev\u002Fnull; then\n    mv \"$tmp\" \"$PATCHES_LEGACY\"\n  else\n    rm -f \"$tmp\"\n  fi\n\n  # Fetch the current directory shape independently. This is the required\n  # fallback when `.printing-press-patches.json` is absent.\n  listing=$(gh api \"repos\u002Fmvanhorn\u002Fprinting-press-library\u002Fcontents\u002F$LIB_PATH\u002F.printing-press-patches\" 2>\u002Fdev\u002Fnull || true)\n  if jq -e 'type == \"array\"' \u003C\u003C\u003C\"$listing\" >\u002Fdev\u002Fnull 2>&1; then\n    mkdir -p \"$PATCHES_DIR\"\n    jq -r '.[] | select(.name | endswith(\".json\")) | \"\\(.name)\\t\\(.download_url)\"' \u003C\u003C\u003C\"$listing\" \\\n    | while IFS=$'\\t' read -r name url; do\n        tmp=$(mktemp)\n        if curl -fsSL \"$url\" -o \"$tmp\" 2>\u002Fdev\u002Fnull; then\n          mv \"$tmp\" \"$PATCHES_DIR\u002F$name\"   # atomic: a dropped transfer never leaves corrupt JSON\n        else\n          rm -f \"$tmp\"\n        fi\n      done\n  fi\nfi\n\n# Count from the first non-empty shape locally; PATCHES_SOURCE is what Phase D reads.\nif [[ -d \"$PATCHES_DIR\" ]]; then\n  DIR_PATCH_COUNT=$(find \"$PATCHES_DIR\" -maxdepth 1 -name '*.json' ! -name '_meta.json' | wc -l | tr -d ' ')\nelse\n  DIR_PATCH_COUNT=0\nfi\nif [[ \"$DIR_PATCH_COUNT\" != \"0\" ]]; then\n  PATCH_COUNT=\"$DIR_PATCH_COUNT\"\n  PATCHES_SOURCE=\"$PATCHES_DIR\"\nelif [[ -f \"$PATCHES_LEGACY\" ]]; then\n  PATCH_COUNT=$(jq '(.patches \u002F\u002F []) | length' \"$PATCHES_LEGACY\" 2>\u002Fdev\u002Fnull || echo 0)\n  PATCHES_SOURCE=\"$PATCHES_LEGACY\"\nelse\n  PATCH_COUNT=0\n  PATCHES_SOURCE=\"$PATCHES_DIR\"\nfi\n```\n\nIf `$PATCH_COUNT == 0` or no index is present (older CLI predating the\npatches contract), skip the rest of this subsection — no patches block in\nthe hand-off.\n\nIf `$PATCH_COUNT > 0`, surface a one-liner to the user before continuing:\n\n> Public `\u003Capi>` has `$PATCH_COUNT` recorded patch(es) against the prior\n> printed CLI. Carry them into the brief as a watch-list. Regen and\n> publish-validate now read the records and fail closed if a recorded\n> file or declared call site \u002F `pp:patch` marker is gone — do not treat\n> the index as a claim that the customization still shipped.\n\nHold `$PATCHES_SOURCE` and `$PATCH_COUNT` for Phase D.\n\n## Phase C — Recency recommendation\n\nPull `researched_at` from the most-recent prior `research.json` and\n`printing_press_version` + `generated_at` from `.printing-press.json`:\n\n```bash\nRESEARCHED_AT=$(jq -r '.researched_at \u002F\u002F empty' \"$MAN_RESEARCH\" 2>\u002Fdev\u002Fnull)\nPRESS_VERSION=$(jq -r '.printing_press_version \u002F\u002F empty' \"$LIB_TARGET\u002F.printing-press.json\" 2>\u002Fdev\u002Fnull)\nGENERATED_AT=$(jq -r '.generated_at \u002F\u002F empty' \"$LIB_TARGET\u002F.printing-press.json\" 2>\u002Fdev\u002Fnull)\n```\n\nCompute the calendar age of the research with `python3` so it stays portable\nacross macOS\u002FLinux and tolerates the microsecond precision that\n`generated_at` carries (BSD `date -f` rejects fractional seconds; `python3`\nis on every supported platform):\n\n```bash\nAGE_DAYS=$(python3 -c \"\nfrom datetime import datetime, timezone\nts = '$RESEARCHED_AT'.replace('Z', '+00:00')\nprint(int((datetime.now(timezone.utc) - datetime.fromisoformat(ts)).total_seconds() \u002F\u002F 86400))\n\" 2>\u002Fdev\u002Fnull)\n```\n\nSurface both signals to the user — research age and prior machine version.\nAge thresholds are rules of thumb, not gates:\n\n- under 30 days → reuse looks safe\n- 30–120 days → reuse plausible; the user should mention any known API\n  churn in their reprint reason so the subagent's Pass 2 picks it up\n- over 120 days → redo recommended\n\nDon't predict API churn from age alone — describe the signals and let the\nuser override. The Phase 0 binary-version-bump revalidation in\n`\u002Fprinting-press` handles the machine-delta side independently; don't\nduplicate it here.\n\nAsk via `AskUserQuestion`:\n\n1. **Reuse prior research** — keep the prior brief; the subagent re-scores\n   prior novel features against current personas\n2. **Redo research** — re-run Phase 1 of `\u002Fprinting-press` from scratch; the subagent still\n   ingests prior novel features as Pass 2(d) input\n3. **Show me first** — display the prior brief's headline + novel-features\n   list, then re-ask between options 1 and 2\n\n## Phase D — Hand off to `\u002Fprinting-press`\n\nBefore invoking `\u002Fprinting-press`, use the prior CLI's scorecard and manifest\nto decide whether the reprint should offer spec enrichment that the first print\ncould not have used. Reprints have better evidence than fresh prints: they know\nwhich structural dimensions were weak, which Printing Press version produced\nthe prior CLI, and what the user named as the reason for regenerating.\n\nFind the most recent scorecard JSON from the prior manuscript run. If no\nscorecard artifact exists, run a fresh structural scorecard against the local\nlibrary copy:\n\n```bash\nSCORECARD_SOURCE=$(ls -1t \"$PRESS_MANUSCRIPTS\u002F$API_SLUG\"\u002F*\u002Fproofs\u002Fscorecard.json 2>\u002Fdev\u002Fnull | head -1)\nSCORECARD_JSON=\"\"\nif [[ -n \"$SCORECARD_SOURCE\" ]]; then\n  SCORECARD_JSON=$(cat \"$SCORECARD_SOURCE\" 2>\u002Fdev\u002Fnull || true)\nelif [[ -d \"$LIB_TARGET\" ]]; then\n  SCORECARD_SOURCE=$(mktemp)\n  if \"$PRINTING_PRESS_BIN\" scorecard --dir \"$LIB_TARGET\" --json > \"$SCORECARD_SOURCE\" 2>\u002Fdev\u002Fnull; then\n    SCORECARD_JSON=$(cat \"$SCORECARD_SOURCE\" 2>\u002Fdev\u002Fnull || true)\n  fi\n  rm -f \"$SCORECARD_SOURCE\"\n  SCORECARD_SOURCE=\"\"\nfi\n```\n\nIf `SCORECARD_JSON` is empty, continue without enrichment prompting and say the\nreprint is proceeding without prior score evidence. Do not invent a prompt from\nthe reprint reason alone.\n\nWhen `SCORECARD_JSON` is available, inspect only dimensions that map to a\npre-generation spec edit and skip dimensions that already score 10\u002F10:\n\n- `mcp_remote_transport`, `mcp_token_efficiency`, `mcp_tool_design`, and\n  `mcp_surface_strategy` below 10 can be lifted by the `\u002Fprinting-press`\n  Phase 2 section **Pre-Generation MCP Enrichment**. Examples:\n  - remote transport is below 10: offer `mcp.transport: [stdio, http]` or the\n    OpenAPI `x-mcp.transport` equivalent before regeneration.\n  - token efficiency, tool design, or surface strategy is below 10: offer the\n    `\u002Fprinting-press` Phase 2 MCP surface decision, including intents for clear multi-step\n    workflows or the Cloudflare pattern for large surfaces.\n- `auth_protocol` below 10, or prior manifest evidence that the CLI used a\n  slug-derived env var where the ecosystem has a canonical env var, can be\n  lifted by **Pre-Generation Auth Enrichment**. Offer to carry canonical\n  `auth.env_vars` or OpenAPI `x-auth-env-vars` guidance into the spec.\n- `data_pipeline_integrity` below 10 is only an enrichment opportunity when the\n  prior CLI or research shows sync-eligible resources. In that case, point the\n  handoff at the relevant `\u002Fprinting-press` Phase 2 sync\u002Fcache enrichment decision rather than\n  treating the score alone as proof that a local store should exist.\n\nUse `AskUserQuestion` for each concrete opportunity before the handoff. Phrase\nthe question around the scorecard evidence and the named canonical section, not\naround a freeform rewrite. Example:\n\n> Prior scorecard shows `mcp_remote_transport: 5\u002F10`. Offer MCP transport\n> enrichment before regenerating, using `\u002Fprinting-press` Phase 2\n> **Pre-Generation MCP Enrichment** as the source of truth?\n\nOptions:\n\n1. **Apply enrichment to the handoff** - include the selected spec edit in the\n   `\u002Fprinting-press` prompt so Phase 2 can update the spec before generation.\n2. **Skip for this reprint** - leave the spec unchanged for this dimension.\n3. **Show score evidence first** - print the relevant scorecard lines and then\n   re-ask between options 1 and 2.\n\nDo not auto-apply enrichment. If the prior scorecard already has\n`mcp_remote_transport: 10\u002F10`, do not ask the redundant MCP transport question.\nIf the user accepts any opportunity, add a `## Reprint Spec Enrichment`\nsection to the `\u002Fprinting-press` handoff. Keep it brief: name the weak\ndimension, the accepted enrichment, and the canonical `\u002Fprinting-press` Phase 2\nsection to execute. Do not duplicate the canonical enrichment text here.\n\nInvoke `\u002Fprinting-press \u003Capi>` and bundle these into the prompt:\n\n1. **A header line** stating the user already chose to regenerate, so\n   `\u002Fprinting-press` Phase 0's library-check should select \"Generate a fresh CLI\" and not\n   re-prompt fresh-vs-improve.\n2. **Research mode** from Phase C (`reuse` or `redo`). `\u002Fprinting-press` Phase 0's existing\n   reuse logic consumes this.\n3. **The user's freeform reprint reason**, verbatim, in a `User context`\n   block. This propagates into the brief as `## User Vision` and becomes\n   Pass 2(e) input to the novel-features subagent — the right hook for\n   \"I want better MCP support\" → bias the brainstorm accordingly.\n4. **Reprint spec enrichment** — only when the scorecard-driven prompt above\n   found an accepted opportunity. Include under a\n   `## Reprint Spec Enrichment` heading.\n5. **Prior patches** — only when Phase B found `$PATCH_COUNT > 0`.\n   Include under a `## Prior Patches` heading.\n\n   Lead the section with this framing sentence, verbatim:\n\n   > The following were hand-fixed against the prior printed CLI. They\n   > are informational — the machine may have absorbed some of these\n   > upstream since the patch was applied. Stay aware so the fresh code\n   > doesn't silently regress the underlying API truth or architectural\n   > pattern, but do not treat as a re-apply checklist.\n\n   Then summarize the patches from `$PATCHES_SOURCE`. Lead each entry\n   with the *substance* the patch encodes (the API truth, the\n   architectural pattern, the cross-file convention) drawn from each\n   patch's `reason` and `validated_outcome` — not just `summary`. Patch\n   metadata (`id`, `files`) is fine as parenthetical context but never\n   the headline.\n\n   Scale the section to patch count:\n\n   - **1–3 patches**: one short paragraph per patch — substance first,\n     then how it manifested. Model the shape on:\n     > Linear's personal API keys go in `Authorization: lin_api_…` raw,\n     > with no `Bearer` prefix; OAuth tokens use `Authorization: Bearer\n     > \u003Ctoken>`. The prior CLI shipped with `auth set-token` only\n     > (Bearer-defaulted) and was patched post-publish to add\n     > `auth set-api-key` plus the raw-header path in `config.go`\n     > (`linear-auth-api-key-vs-oauth-token`).\n   - **4–9 patches**: one tight sentence per patch, same substance-first\n     shape.\n   - **10+ patches**: thematic summary by category (auth, pagination,\n     query encoding, MCP additions, helpers, classifiers, error\n     envelopes). Two or three sentences per theme citing 2–3 patch `id`s\n     as evidence. Do not enumerate all entries inline.\n\n   Close the section with a pointer so a downstream agent can drill into\n   any specific patch when working in the relevant area:\n\n   > Full patch detail: the per-patch files under `$PATCHES_SOURCE`\n   > (each `\u003Cid>.json` is one self-contained patch; the legacy\n   > single-array `.printing-press-patches.json` carries the same fields\n   > under `patches[]`).\n\nDo **not** pass a separate \"this is a reprint\" marker. The novel-features\nsubagent runs unconditionally on every print and discovers prior research\nvia its own discovery snippet. The paths\nimport populated in Phase A are exactly the paths it checks; Pass 2(d)\nfires whenever prior `research.json` exists.\n\nThe library-preservation contract is owned by `\u002Fprinting-press` Phase 5.6\n(\"Promote to Library\"), not by this skill. When the existing library has\n`novel_features > 0` in its manifest (or hand-authored files under\n`internal\u002Fcli\u002F`, `internal\u002Fsyncer\u002F`, or `internal\u002Fstore\u002F`), Phase 5.6 first\ndry-runs `\"$PRINTING_PRESS_BIN\" regen-merge \"$LIB_TARGET\" --fresh \"$CLI_WORK_DIR\"\n--json` to decide whether the fresh tree rebuilt the prior novels. If the\nfresh tree contains all prior novel work, Phase 5.6 uses the swap path and\ntreats generated-file version drift as expected overwrite. Otherwise it routes\npromotion through `regen-merge --apply` so still-unique hand-authored novels\nsurvive the reprint and genuine `NOVEL-COLLISION` \u002F missing-referent cases halt\nfor review. This honors the prefer-`regen-merge` guidance under the\n**Hand-edits must be regen-mergeable.** section of\n`\u002Fprinting-press` Phase 3. If a future\nedit to that phase changes the routing rule, update this paragraph in the same\nPR -- the reprint skill is the dominant entry point that fires it.\n\nBefore the hand-off, compare regenerated manifest files against the tracked\npublished tree. If `$LIB_TARGET\u002Fmanifest.json` or\n`$LIB_TARGET\u002Ftools-manifest.json` exists, diff each file against the freshly\ngenerated counterpart under `$CLI_WORK_DIR` and surface non-empty diffs in the\nhandoff prompt. Treat a diff as a reconciliation checkpoint, not as automatic\noverwrite approval: the operator must decide whether to preserve the tracked\nhand-edit, fold it into the spec\u002Fresearch inputs, or intentionally accept the\nregenerated value. Do not continue silently when tracked manifest fields,\ntool metadata, or descriptions would be dropped.\n\nAttribution also stays owned by `\u002Fprinting-press`: the hand-off runs generation\nfor the same API slug, and the generate\u002Fpromote path must preserve the existing\nlibrary manifest's permanent `creator` while adding the reprinter to\n`contributors[]` when they differ. Do not repair this by hand-editing\n`creator`, `contributors[]`, README bylines, SKILL `author:`, or NOTICE; rerun\nwith a current Printing Press binary so the manifest-first guard rewrites the\nworking tree and promoted tree consistently.\n\n## After hand-off\n\nThe printing-press flow drives the rest. Don't summarize its work — let\nthe user see the live phases.\n\nIf `\u002Fprinting-press` halts with the subagent's pre-flight HALT (brief\nlacks concrete `Users` \u002F `Top Workflows` content), the reused prior brief\npredates the subagent's required schema. Recommend re-running with **Redo\nresearch** selected at Phase C.\n",{"name":5,"created_by":25,"description":26,"allowed-tools":27},"user","Regenerate an existing printed CLI from scratch under the current Printing Press, with prior research, prior novel features, and prior patches (post-publish hand-fixes) carried into the writing pipeline as reconciliation context rather than dropped on the floor. Pulls the CLI from the public library if it isn't local, recommends reuse-vs-redo of prior research based on age, then hands off to \u002Fprinting-press with the right context. Use when a machine upgrade would benefit a published CLI more than manual polish. Trigger phrases: \"reprint \u003Capi>\", \"regenerate \u003Capi>\", \"redo the \u003Capi> CLI\", \"rebuild \u003Capi> from scratch\", \"this CLI would benefit from a reprint\".\n",[28,29,30,31],"Bash","Read","AskUserQuestion","Skill","main","e63612d78f9099c1970584ea3faef109f42d20bb","skills\u002Fprinting-press-reprint",0,"https:\u002F\u002Fgithub.com\u002Ffilippolmt\u002Fskills","https:\u002F\u002Fgithub.com\u002Ffilippolmt","2026-11-14 22:46:00",{"source":8,"skillName":5,"commands":40,"prompt":50,"zipAvailable":51},[41,44,47],{"tool":42,"command":43},"npx","npx skills add filippolmt\u002Fskills --skill printing-press-reprint",{"tool":45,"command":46},"bunx","bunx skills add filippolmt\u002Fskills --skill printing-press-reprint",{"tool":48,"command":49},"pnpm","pnpm dlx skills add filippolmt\u002Fskills --skill printing-press-reprint","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Ffilippolmt\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fprinting-press-reprint\u002FSKILL.md",true,[53,57],{"path":54,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":56,"sourceHtmlUrl":16},"LICENSE","许可证",1086,{"path":58,"resourceType":59,"resourceTypeDesc":60,"sizeBytes":61,"sourceHtmlUrl":62},"SOURCE.md","REFERENCE","参考文档",426,"https:\u002F\u002Fgithub.com\u002Ffilippolmt\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fprinting-press-reprint\u002FSOURCE.md",[64,72,81,89,97,105],{"skillNo":65,"name":66,"nameZh":6,"description":67,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":68,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":20},"01M781ZJN9KDK0YY8EAQ8WXPRQ","writing-for-agents","Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.","https:\u002F\u002Fgithub.com\u002Ffilippolmt\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwriting-for-agents\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Ffilippolmt\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwriting-for-agents\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":76,"licenseFileUrl":77,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":20},"01M781ZDZ8CQ30GRY5XNT1RF91","wrangler","Run or troubleshoot Wrangler CLI commands in projects that have a wrangler.jsonc, wrangler.json, or wrangler.toml file, including local development, Previews, deployment, and resource management. For new projects and projects without a Wrangler configuration file, use the cf CLI instead.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Ffilippolmt\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwrangler\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Ffilippolmt\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwrangler\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":76,"licenseFileUrl":85,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":20},"01M781ZA1MB7E7T1R1TE0BTTJS","workers-profiling","Profile or debug CPU usage and memory allocation in deployed Cloudflare Workers and Durable Objects. Then use this information to optimize your code.","https:\u002F\u002Fgithub.com\u002Ffilippolmt\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fworkers-profiling\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Ffilippolmt\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fworkers-profiling\u002FSKILL.md",[],[],{"skillNo":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":76,"licenseFileUrl":93,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":20},"01M781Z52Q7PBMDA2YBC3SCTYM","workers-development","Run, inspect, and debug Cloudflare Workers during local development with cf dev, wrangler dev, or the Cloudflare Vite plugin, using Local Explorer to read and seed binding data and query local traces and logs. Also routes production debugging to Workers observability. Use for debugging requests whether or not the project uses Wrangler.","https:\u002F\u002Fgithub.com\u002Ffilippolmt\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fworkers-development\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Ffilippolmt\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fworkers-development\u002FSKILL.md",[],[],{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":76,"licenseFileUrl":101,"sourceHtmlUrl":102,"categories":103,"keywords":104,"publishedAt":20},"01M781Z193MG7SHF4DSD1WK583","workers-best-practices","Cloudflare Workers best practices for production applications. Use when writing, reviewing, or configuring Workers.","https:\u002F\u002Fgithub.com\u002Ffilippolmt\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fworkers-best-practices\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Ffilippolmt\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fworkers-best-practices\u002FSKILL.md",[],[],{"skillNo":106,"name":107,"nameZh":6,"description":108,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":109,"sourceHtmlUrl":110,"categories":111,"keywords":112,"publishedAt":20},"01M781YT0JB3GVA1HTFZ9QJHM7","wordpress-pro","Develops custom WordPress themes and plugins, creates and registers Gutenberg blocks and block patterns, configures WooCommerce stores, implements WordPress REST API endpoints, applies security hardening (nonces, sanitization, escaping, capability checks), and optimizes performance through caching and query tuning. Use when building WordPress themes, writing plugins, customizing Gutenberg blocks, extending WooCommerce, working with ACF, using the WordPress REST API, applying hooks and filters, or improving WordPress performance and security.","https:\u002F\u002Fgithub.com\u002Ffilippolmt\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwordpress-pro\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Ffilippolmt\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwordpress-pro\u002FSKILL.md",[],[]]