skillZs
★ LIVE SKILL TAGS ★
>>> LIVE SKILLS INDEX <<<
* OPEN SOURCE *
NO LOGIN, NO TRACKING
※ REAL INSTALL DATA ※
← back to all skills
shengchen1998/author-toolkit49 installs

revise-increment

Bring added or affected paths to current-draft revision convergence for the target scope while reusing full local coverage units from unchanged paths.

How do I install this agent skill?

npx skills add https://github.com/shengchen1998/author-toolkit --skill revise-increment
view source ↗

Is this agent skill safe to install?

  • Gen Agent Trust Hubpass

    The skill is a workflow coordinator for incremental manuscript revisions within the Author Toolkit ecosystem. It operates entirely on local files and sibling components from the same vendor using relative paths. No malicious patterns, exfiltration, or unauthorized remote access were detected.

  • Socketpass

    No alerts

  • Snykpass

    Risk: LOW · No issues

What does this agent skill do?

Revise Increment

Coordinate revision after incremental manuscript growth. Use the current-draft operations and shared evidence to revise added or affected paths without treating creation acceptance as current-draft revision convergence. One manifest continues to represent one book. Toolkit skills from one installation are siblings. Resolve <setup-skill-root> as ../setup-author-toolkit relative to this skill's installed directory and use the controlled workflow terminology in <setup-skill-root>/assets/references/terminology.md for shared terms and exact contract values.

Start or resume

  1. Locate .author-toolkit/project.json. If it is absent, display the host command for setup-author-toolkit and stop. Use the same setup command forms as revise-fiction. If the manifest has any schema version other than 3.1.0, report that it is unsupported and stop without changing it.
  2. Read each needed callee's installed SKILL.md and applicable references. Execute its procedure directly. A user-invoked entry starts only through its author-facing command, never as an internal callee.
  3. Run the author-project procedure once with check --digest, passing --require-sibling author-project --require-sibling revise-fiction --require-sibling revise-draft. Its single JSON result supplies the bounded projection, project validation, ordinary status, and refreshed navigation cache, with installation and entry-file checks included; use this result instead of separate doctor, read, validate, status, or standalone digest-generation calls for this startup. If it reports an incomplete toolkit installation, stop and report the missing installed files; do not deploy contracts into the project. If another check fails, rerun it once with --diagnose instead of --digest, report the failure, and stop at the applicable author gate. Read PROJECT_STATE.md first, then read only the tracker sections, sessions, current plan and scope, pending action, applicable creation batches and commit journals, coverage units and index, review jobs, and prior revision evidence needed for the requested work. Do not read the full tracker or backup index for routine navigation. For routine navigation, use PROJECT_STATE.md pointers, targeted artifact reads, or aggregate counts rather than emitting complete file listings of the registered session or backup directories. Reconcile an interrupted write or commit before dependent work.
  4. On a new increment, establish the previous scope, the requested target scope, and the change set. Verify that the previous scope has reached current-draft revision convergence, including author-project status with the previous scope ID and --require-execution; a recorded statement of that state does not replace this check. If it has open findings, unresolved required coverage, or unverified approvals, show the revise-draft command and stop rather than skipping unfinished work for the previous scope. Record the affected operation as blocked when strict execution status fails. The target scope normally contains every retained chapter from the previous scope plus the selected new chapters. The change set includes every created, modified, deleted, or reordered path and every path whose evidence or dependencies changed. Creation batches can locate changes; their create action alone cannot prove that other chapters are unaffected.
  5. On resume, locate the active increment in the latest session and any current plan. Treat its verified previous scope as historical evidence: the increment's own approved edits may stale that previous scope without reopening the entry gate. Rederive the current change set, work set, and current-draft revision convergence for the target scope from current evidence instead of requiring the edited manuscript to match the previous scope. Reconcile external manuscript changes and any interrupted write or commit before dependent work. If no active increment with a verified previous scope can be established, apply the new-increment entry gate.
  6. Derive the work set from target-scope paths that need new local evidence or edits. Include changed paths, paths with stale dependencies, and paths without a reusable full local coverage unit for a required operation. Unchanged paths remain reusable only when operation, source-report identity, paths and hashes, requested subchecks, and dependencies match. Whole-report evidence for the previous scope does not establish coverage for the expanded target scope.
  7. On the first run, select intake. On later runs, select the earliest missing, partial, or stale evidence in the sequence below, including review findings that require another revision pass.
  8. Record the selected operation through author-project with set-pending. Run one operation and stop at its author gate with awaiting_author or blocked. For intake, complete the read-only inventory and present the previous scope, proposed or confirmed target scope, change set, work set, reusable evidence, and evidence gaps. Stop for author confirmation of the target scope and work set. If the confirmed current scope already matches the requested target scope, reuse it as the target scope instead of manufacturing a replacement; otherwise use author-project with record-scope to record the proposed target before presenting it, then confirm that exact unchanged record after explicit author confirmation. Target-scope and work-set confirmation do not approve manuscript edits.

The operation sequence is intake, draft-baseline, diagnostics, characters, revision-plan, chapter-revision, prose, ai-cleanup, and current-scope-review. Follow the corresponding gates in ../revise-draft/references/stage-gates.md, with the incremental rules below. Clean-audit continuation within prose follows prose-mechanics and does not advance to another operation.

Incremental operation routing

For each diagnostic callee, supply the operation identifier, destination report path, named inputs and versions, target scope_id, BCP 47 language, tracker path, and resolved <setup-skill-root>. Require <setup-skill-root>/assets/references/workflow-reporting.md and finding v2, and supply ../revise-fiction/references/state-and-findings.md resolved from the callee's installed skill directory as the shared state reference. Identify the work-set paths in the report notes and record actual operation coverage and subchecks.

OperationIncremental work
intakeInventory the complete manuscript; establish the previous scope, target scope, change set, work set, and reusable evidence
draft-baselineUpdate the baseline for the full target scope from current text; reuse labeled prior observations and show affected differences
diagnosticsDiagnose every work-set chapter and its transitions and dependencies; compose current local coverage for the full target scope
charactersRecheck affected current arcs, continuity, knowledge states, timeline, and world facts on work-set paths; compose character coverage for the full target scope
revision-planConsolidate current accepted findings for the target scope into exact work-set batches; keep no-edit paths and deferred decisions visible
chapter-revisionApply only approved work-set batches with current hashes and backups; expand the work set when an accepted fix affects another path
proseRun requested audits on work-set text; compose matching full local coverage units and a new full-target synthesis
ai-cleanupRun requested detection or approved rewrites on work-set text; compose language coverage and operation coverage for the full target scope
current-scope-reviewRun author-review with review_mode=current-scope over the full confirmed target scope and all three perspectives

For diagnostics, characters, prose, and AI cleanup, persist full local coverage units for independently covered paths and compose the living coverage index for the target scope. During sequential prose audits, retain every accumulated matching unit when replacing the prose entry; keep synthesis partial or unrecorded until the selected queue finishes. Write a new full-target synthesis whenever membership, order, text, dependencies, or child coverage changes. Missing reusable units add the affected old paths to the work set; never relabel old whole-report evidence as target-scope evidence.

The full-target current-scope review has the same scope and actual-reading requirements as revise-draft. In particular, the fresh reader must read the full target scope sequentially through one continuous reviewer state. Prior reports from the previous scope do not replace review units or syntheses for the target scope. Limiting edits to the work set does not limit the review's reading scope.

Current-draft revision convergence loop

Use the shared Current-draft revision convergence definition at every completion check.

After each full-target current-scope review, resolve every returned finding through the author decision loop. An accepted edit returns to revision-plan and chapter-revision, stales affected downstream evidence, and requires new affected coverage and another full-target current-scope review. Deferred or dismissed findings retain their explicit decisions. Creation acceptance, a completed chapter-review, one clean diagnostic, or a review with undecided findings does not establish current-draft revision convergence.

Claim current-draft revision convergence only when every required operation is current for the target scope, all requested local coverage and syntheses are complete, every finding has a current decision, no approved batch remains unexecuted, and the full-target current-scope review satisfies its gate after the latest edits. Record the no-edit skip for revision-plan and chapter-revision when applicable. Keep partial, skipped, or degraded requested applicable coverage visible as a blocker unless the author explicitly narrows the requested pass; name that narrowed requested pass instead of claiming current-draft revision convergence. Documented inapplicable whole-book or language-specific checks may remain explicitly skipped with their reasons.

Author interaction and manuscript safety

At every author gate, follow the Author decision loop. Before manuscript changes, read <setup-skill-root>/assets/references/context-fit.md and complete its check before presenting an edit proposal. Show the exact operation, target scope, work-set paths, actions, input versions, exact candidates, and checked local context. Record explicit approval in the shared tracker, then run author-project with record-backup for every affected file and approval ID. Run complete-write with the exact approved content to publish it, preserve evidence, and record the execution receipt in one invocation; an interrupted batch has no execution receipt until reconciliation and remains unverified. Restore requires separate explicit approval and a safety backup.

Strategy switching and shared state

write-fiction, revise-draft, revise-fiction, revise-increment, and revise-request use the same manifest, tracker, reports, sessions, backups, handoffs, and creation state. At any stable operation gate, show the exact command for the author's selected strategy and wait. Switching does not reset evidence, resume writes, or recreate shared project state. This entry adds no report v2 stage value, pending-action operation name, manifest field, or separate state tree.

Follow Task-based reading before writing a report, changing finding or approval status, propagating stale evidence, or recording a stopping point. Validate newly written reports and artifacts with the installed validator; preserve historical evidence by default or retire only unreferenced audit artifacts under the explicitly enabled bounded policy.

Finish

Finish each invocation with the previous-scope and target-scope IDs, change-set and work-set paths, reused full local coverage units, new reports and syntheses, actual coverage, approvals and backups, unresolved findings, and the exact next action. At the final gate, state explicitly whether the target scope has reached current-draft revision convergence and list every blocker when it has not. Show the write-fiction, revise-draft, revise-fiction, revise-increment, or revise-request command that matches the author's next goal. Use author-handoff for session transfer after shared state is current.

Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.

<a href="https://skillzs.dev/skills/shengchen1998/author-toolkit/revise-increment">View revise-increment on skillZs</a>