Skip to content
Agentmart
Submit a tool

pac-deep-read

Analyze substantial documents beyond summary by inferring implications, surfacing tensions, extracting a core takeaway, and identifying blind spots.

Skill by mypac

IndexedMIT★ 2
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/ladislas-pac-deep-read

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Analyze substantial documents beyond summary by inferring implications, surfacing tensions, extracting a core takeaway, and identifying blind spots. Use when the user explicitly asks for a deep or critical reading, hidden implications, tensions, contradictions, or blind spots—not for an ordinary summary.

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found
Runs as
Node.js package

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/pac-deep-read

What's inside

Skill name
pac-deep-read
Description
Analyze substantial documents beyond summary by inferring implications, surfacing tensions, extracting a core takeaway, and identifying blind spots. Use when the user explicitly asks for a deep or critical reading, hidden implications, tensions, contradictions, or blind spots—not for an ordinary summary.
License
MIT
Metadata
stage: shared · author: mypac
Instructions
75 lines in SKILL.md · view on GitHub

Before installing

  • Ships executable scripts

Files (2)

  • SKILL.md
  • skill.test.mjs

SKILL.md (first lines)

# Deep-read a document

Analyze substantial written material beyond summary. Work from document content already available in the conversation or supplied by the user; if the target is missing or ambiguous, ask for it rather than inventing content.

Use one of two modes:

- **Concise mode (default):** use unless the user requests long, rigorous, evidence-backed, or detailed analysis.
- **Rigorous / evidence mode:** use when the user makes one of those requests or explicitly selects rigorous mode.

Infer carefully. Distinguish the author's claims from your inferences, ground conclusions in the source, and do not manufacture entries for categories that genuinely do not apply.

## Concise mode

Use these sections in order, keeping each to roughly 2–3 sentences:

1. **Non-obvious insights (inferred):** identify 3–5 implications not explicitly stated by the author.
2. **Tensions / contradictions:** surface internal tensions, unresolved tradeoffs, or places where the argument sits uneasily with conventional wisdom.
3. **Core takeaway:** give one actionable implication—or one important reframing for a theoretical text—and explain why it matters.
4. **Blind spots:** identify external perspectives, data, or questions the author should engage with but does not.

If a category genuinely does not apply, say so briefly.

## Rigorous / evidence mode

Write concise bullets with one sentence per bullet. Provide 3–5 Insights, 2–4 Tensions, exactly 1 Takeaway, and 2–4 Blind spots unless a category genuinely does not apply; in that case, say so briefly rather than manufacturing an entry.

Use this output contract:

### 1) Non-obvious insights (inferred)

For every Insight include:

- **Insight:** an inference not explicitly stated by the author.
- **Evidence:** a short verbatim excerpt from the document, approximately 5–12 words.
- **Assumption:** the key assumption required for the inference.
- **Confidence:** High / Med / Low.

### 2) Tensions / contradictions

For every Tension include:

- **Tension:** the conflict, tradeoff, or unresolved point.
- **Evidence:** a short verbatim excerpt from the document, approximately 5–12 words.
- **Why it conflicts / what remains unresolved:** the source of the tension.

### 3) Core takeaway

Include:

- **Takeaway:** the single most important actionable implication, or reframing for a theoretical text.
- **Why it matters:** its significance.
- **Suggested next step:** a proportionate follow-up.

### 4) Blind spots

For every Blind spot include:

- **Blind spot:** the missing perspective, data, or question.
- **Why it matters:** the consequence of the omission.
- **What would resolve it (data / perspective / question):** the evidence or inquiry needed.

### 5) Next action

Include this section only when the provided material contains enough organizational context to identify a meaningful action, owner, output, and success criteria. Omit it when organizational context is absent.

- **Action:** the concrete action.
- **Owner (role/name):** the responsible role or person.
- **Output expected (artifact/decision):** the deliverable.
- **Success criteria:** the observable completion condition.

## Example

A request to “deep-read this strategy memo” uses concise mode; “give me a long, evidence-backed deep read of this strategy memo” uses rigorous / evidence mode.

From ladislas/mypac, MIT.

More from ladislas/mypac

Signals

Stars
2
Forks
1
Open issues
3
Last commit (this folder)
Oct 15, 2026 (14 days ago)
Last release
Oct 15, 2026 (14 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from mypac

Cite this page

pac-deep-read skill. Agent Mart. https://agentmart.dev/skills/ladislas-pac-deep-read (last updated 2026-10-24)