تشغيل أي مهارة في Manus بنقرة واحدة

doc-review

النجوم١

التفرعات١

آخر تحديث١٧ مايو ٢٠٢٦ في ١٣:٢٧

In-browser annotation overlay for HTML, Markdown, and plain-text documents. Use when a user wants to annotate a document, run a structured review pass with severities (typo / edit / rewrite), or apply review comments back into source files via Claude Code.

التثبيت

التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.

تشغيل في Manus

المصدر

Entelligentsia

Entelligentsia/skillforge

فتح مستودع GitHub عرض مستودعات المنشئ

تنزيل

تشغيل في Manus

المهن ذات الصلةSOC

استنادا إلى تصنيف SOC المهني

مدقّقو النصوص ومحرّروهادعم المكاتب والإدارة·SOC 43-9081

SKILL.md

readonly

name	doc-review
description	In-browser annotation overlay for HTML, Markdown, and plain-text documents. Use when a user wants to annotate a document, run a structured review pass with severities (typo / edit / rewrite), or apply review comments back into source files via Claude Code.

doc-review

Action

This skill is a delegator. When invoked, immediately route to the right slash command based on user intent:

File argument present (e.g. @README.md or a path to .html / .md / .markdown / .txt / .rst) AND user wants to start or resume an annotation session → invoke /doc-review:review-doc <file> (pass through any --port N).
User says "apply review" / "apply comments" / mentions a .review.json → invoke /doc-review:apply-review [<json-path>].
User says "stop" / "kill review" → invoke /doc-review:review-doc-stop.
User mentions polling, cron tick, internal loop → these run automatically; do not invoke /doc-review:apply-review-tick manually.

Do NOT print the documentation below as the response. Print at most a one-line confirmation of what command you are routing to, then invoke it.

Reference (for Claude's own context)

Workflow summary:

A Node HTTP server serves the document in a browser with an injected sidebar.
A reviewer adds anchored comments (selected text + 30-char context).
Each comment carries a severity: typo, edit, or rewrite.
Clicking Apply on a comment in the browser queues it; Claude Code, driven by a /loop tick command, picks it up within ~60 seconds and edits the source file in place.
The server is multi-session aware: two Claude sessions reviewing the same doc share one server; a server stays up while at least one session holds a reference, and self-terminates when the last reference drops.

Commands:

/doc-review:review-doc <file> [--port N]
/doc-review:review-doc-stop
/doc-review:apply-review [<path-to-review.json>]
/doc-review:apply-review-tick --port <N> --session-id <id> (loop-only)

Storage layout:

Per-project: <project-root>/.doc-review/<slug>.review.json, .../<slug>.queue.json, .../.servers/<slug>.json.
Per-session: $HOME/.cache/doc-review/sessions/<session_id>.json.

Project root detected by walking up from the doc for .git/, .doc-review/, package.json, pyproject.toml, Cargo.toml, go.mod, .hg/, or .svn/. Override with --project-root.

المزيد من هذا المستودع

نفس المستودع

claude-usage

Entelligentsia/skillforge

Compute Claude Code token usage by model for any date range from local session transcripts. Use this whenever the user asks how many tokens they used, their Claude Code / Claude subscription usage, usage for a month or week or day, a "usage report", per-model token breakdown, or "how much did I spend/use" over some period — including phrasings like "first week of June", "last 7 days", "June 2026", or "yesterday". This is the right tool on the subscription plan, where console.anthropic.com does NOT show usage. Trigger even when the user just names a period and the word "usage" without saying "tokens".

2026-07-081

walkinto

Entelligentsia/skillforge

WalkInto.in virtual tour platform and helpdesk — authenticate, view profile, manage 360 tours, search the knowledge base, and run the support ticket lifecycle. Use this skill whenever the user mentions 'walkinto', a WalkInto account, virtual/360 tours, panorama uploads, tour analytics, OR the WalkInto helpdesk: support tickets, customer questions, knowledge-base/KB articles, replying to or triaging tickets, ticket status/priority/tags, or 'who am I on walkinto'.

2026-06-251

lean-ctx-eval

Entelligentsia/skillforge

Evaluate whether the lean-ctx MCP is saving more context tokens than it costs, for the current project, using that project's Claude Code session transcripts. Measures ctx_read with a per-call counterfactual vs native Read, compares ctx_shell against native Bash observationally, nets out the per-session instruction+schema tax, and reports a token ledger. Use when asked to eval lean-ctx, measure context-compression savings, audit token usage of an MCP, or decide if lean-ctx earns its seat.

2026-06-021

dev-tooling-audit

Entelligentsia/skillforge

Inspect a project and recommend developer tooling (LSP, AST, test runners, debuggers, linters) that would make Claude Code most effective. Provides exact install commands and a verify mode to confirm each tool works.

2026-05-261

concurrency-control

Entelligentsia/skillforge

Use when a model emits multiple tool calls in a single turn and the harness must execute them in parallel safely, or when subagents/forks run concurrently and share state. Apply when designing the concurrency model for tool dispatch. Covers parallel tool execution, ordering guarantees, cancellation propagation, race detection on shared resources (filesystem, db), partial-failure semantics, and result aggregation.

2026-05-061

hook-system

Entelligentsia/skillforge

Use when an agent harness needs user- or admin-configurable interception points — before/after tool calls, on prompt submit, on stop, on session start. Apply when designing the event/hook architecture that lets operators inject policy, logging, or transformation without modifying core harness code. Covers event taxonomy, blocking vs non-blocking hooks, hook contract (stdin/stdout/exit-code), failure modes, and security implications.

2026-05-061

doc-review

Action

This skill is a delegator. When invoked, immediately route to the right slash command based on user intent:

File argument present (e.g. @README.md or a path to .html / .md / .markdown / .txt / .rst) AND user wants to start or resume an annotation session → invoke /doc-review:review-doc <file> (pass through any --port N).

User says "apply review" / "apply comments" / mentions a .review.json → invoke /doc-review:apply-review [<json-path>].

User says "stop" / "kill review" → invoke /doc-review:review-doc-stop.

User mentions polling, cron tick, internal loop → these run automatically; do not invoke /doc-review:apply-review-tick manually.

Do NOT print the documentation below as the response. Print at most a one-line confirmation of what command you are routing to, then invoke it.

Reference (for Claude's own context)

Workflow summary:

A Node HTTP server serves the document in a browser with an injected sidebar.

A reviewer adds anchored comments (selected text + 30-char context).

Each comment carries a severity: typo, edit, or rewrite.

Clicking Apply on a comment in the browser queues it; Claude Code, driven by a /loop tick command, picks it up within ~60 seconds and edits the source file in place.

The server is multi-session aware: two Claude sessions reviewing the same doc share one server; a server stays up while at least one session holds a reference, and self-terminates when the last reference drops.

Commands:

/doc-review:review-doc <file> [--port N]

/doc-review:review-doc-stop

/doc-review:apply-review [<path-to-review.json>]

/doc-review:apply-review-tick --port <N> --session-id <id> (loop-only)

Storage layout:

Per-project: <project-root>/.doc-review/<slug>.review.json, .../<slug>.queue.json, .../.servers/<slug>.json.

Per-session: $HOME/.cache/doc-review/sessions/<session_id>.json.

Project root detected by walking up from the doc for .git/, .doc-review/, package.json, pyproject.toml, Cargo.toml, go.mod, .hg/, or .svn/. Override with --project-root.