Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Superstack Digest

ASecurity

A period report for a reader outside the session — a weekly update, client status, "what did we get done this month". Skip for end-of-task reports (superstack-ship), retros when a retro tool ships, and recalling past sessions for yourself (a memory tool).

2 stars
0 votes
0 copies
1 views
Added 9/19/2026
ai-agentsgogit

Works with

cli

Security Analysis

A100/100

Scanned 9/19/2026

Install to Claude Code

$npx -y skills add debabsah/superstack --skill superstack-digest --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Superstack Digest?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Superstack Digest
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/debabsah-superstack-digest/badge)](https://www.skillsdirectory.com/skills/debabsah-superstack-digest)

More formats (shields.io, HTML) on the badges page.

Download with Pro
Files
SKILL.md
---
name: superstack-digest
description: A period report for a reader outside the session — a weekly update, client status, "what did we get done this month". Skip for end-of-task reports (superstack-ship), retros when a retro tool ships, and recalling past sessions for yourself (a memory tool).
---

# superstack-digest

The digest is the calibration record paying rent outward. It is assembled from records, never from memory — the conversation is scratch; files are the project (superstack-continuity's rule, applied to reporting).

The description's deferrals (a retro tool for retrospectives; a memory/session-history tool for self-recall) bind only if the named tool is really present — check the session's skill listing first and announce the hand-off; when nothing is there, assemble from records right here.

## Assemble

1. **Gather the period's facts from the record:** `git log` across the repos in scope, the claims-log's `Verified:` lines, value-log settlements (`HELD`/`MISSED`), queue movements, open residuals and skipped-gates. Your recall of the period is a hypothesis; the logs are the evidence (R1 on your own memory). **If the record is thin** — early days, files not yet grown — say so: build from `git log` alone and ledger the gap (`Assumed: the period wasn't fully logged — built from git history only`).
2. **Numbers in a digest are data claims.** Sanity-check each before quoting — counts from commands, reconciled against a known total where one exists. A `PROVISIONAL` number never ships to an external reader: round it into honest words or leave it out.

## Write for the named reader

3. **Their language, not the method's.** Shipped work in outcome words, not commit words ("the report now runs in under a minute", not "refactored the aggregation pipeline"). No session jargon, no tier or ledger tokens **in the artifact** — the draft the client reads goes tokenless. **The chat turn that offers the draft is a normal report and carries its ledger** (`Verified:` what you assembled and from which logs, `Assumed:` what the record couldn't show) — the claims gate reads your turn, not the attachment, and the grammar leaves the artifact, never the method. The honesty travels into the artifact anyway: "on track" is a claim; write it only when the record supports it, and never soften a real problem (R5 holds even in business register — *especially* there).
4. **The reader's sections:** done · in progress with expected next · **blocked / asks** (what you need from *them*, explicit — a digest that hides its asks is decorative) · risks worth their attention.
5. **End the same way every time:** where things stand, what happens next and whose move it is.

## The boundary

**Draft, don't send.** Write the draft to a file named for the period (`<period>-digest.md`, or where the human asks) — outward-facing, so sending is the human's act (the inward/outward line). Offer it with anything you'd flag before it travels — a number you couldn't verify, a claim you'd want them to gut-check — stated above the fold, not discovered by the client.

Attribution

debabsahdebabsah
View sourceMore from debabsah →
SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Related Skills

Caveman

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

1066601 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

686011 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3351 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

651 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →