kz-tob
4b1b74b181
* Give differential-review a trigger, and name every component in its README differential-review's description listed what it does and never named a situation, so it competed on capability wording alone. It now closes with the triggers its own README already documents — reviewing a PR, commit, or diff; checking whether a change re-introduces a fixed bug; asking what else a change could break; finding modified code with no test. The same plugin's README never mentioned adversarial-modeler, which is what Phase 5 dispatches for HIGH RISK changes. Checking whether that was isolated turned up more of it, and the sweep found three kinds of gap: zeroize-audit's agent table was missing three of its eleven agents — 0-preflight, which gates the entire run, plus 5b-poc-validator and 5c-poc-verifier. All three appear in the phase diagram directly above the table, which is why they read as present. constant-time-analysis documents the ct-analyzer CLI end to end and never says the plugin also ships a skill and a command. entry-point-analyzer lists phrases that trigger its skill but never names the skill or its command. Three more READMEs describe their skill without naming it. That matters most where the skill name is not the plugin name and a user cannot guess it: chrome-mcp-troubleshooting and interpreting-culture-index. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * Fix review findings and make the README sweep a gate The two PoC rows I added to zeroize-audit said Phase 4. The diagram three lines above them, SKILL.md, and workflows/phase-5-poc-validation.md all say Phase 5, steps 5a and 5b. "Wave 5a" is a label that exists nowhere. A debugger consulting the table — the artifact this branch designates as what runs when — would have opened phase-4-poc-generation.md and found no validation in it. Also corrected the sentence introducing that table, which still said 10 agents across 8 phases against 11 across 9, and the Phase 0 diagram line, which still credited the orchestrator for a gate the new row credits to 0-preflight. differential-review's README claimed the agent is "dispatched", and named it bare in a column whose other rows are namespaced. Nothing dispatches it: the only instruction is prose in SKILL.md, and a bare subagent_type fails at runtime. Namespaced both, and corrected the five stale line counts in the same file — reporting.md is 369 lines, not the ~120 the token-efficiency section budgets for. Drop the dead `name: trailofbits:<cmd>` key from five command files. The three newest command files carry no name: at all, #275 namespaced 22 bare invocations, and this branch documents the `/<plugin>:<cmd>` form — so the key contradicts the docs it sits next to. Then make the sweep repeatable. Doing this by hand three times found eight gaps and missed two more, both of the same shape: a workflow ships under meta.name, not its filename, so a README citing the filename never writes the name a reader types. The validator now checks that a README names every skill, agent, command, and workflow its plugin ships, reading meta.name for workflows. It refuses a run that inspected zero components, and six self-test assertions hold it to known-bad fixtures. It found git-cleanup on its first run: ships as /git-cleanup:git-cleanup-analysis, README cites workflows/analyze-branches.js four times and that name never. static-analysis had the same gap for codeql-build. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * Fix both P2s: the gate was a substring test, and the dispatch was still bare The README gate ran `name not in text`. That reads as thorough and could not fail for a large share of what it counted: `draw` was satisfied by "(draw cards instead)", `semgrep-rule` by the plugin's own name in the install line, `burp-search` by a `scripts/burp-search.sh` path that is a different thing, and `audit` by the prose "shared-state struct audit". Match by kind instead. Commands and workflows are reachable only as `/<plugin>:<name>`, so require that literal — it is the only string a user can type. Agents are dispatched by identifier and never typed as prose, so require an identifier-shaped mention. Skills are genuinely referred to by bare name, so require only a delimited occurrence, which is what stops "draws" counting as `draw`. That surfaced seven real gaps, the four above plus insecure-defaults' audit-pipeline workflow, mutation-testing's skill, and trailmark's code-slice-worker. All seven fixed. adversarial-modeler was still bare at SKILL.md:96. Line 77 was the decision-tree mention; line 96 is the "Delegate to this agent" instruction a model actually acts on, so the runtime failure the last commit claimed to fix survived it. Namespaced, and it now says why. Also from the review: a per-kind floor, since a single total stays healthy while skill_files() — 63% of coverage — silently stops matching; workflow_names anchored to the meta block, because a bare search takes any earlier `name:` in a comment, and .mjs was invisible; and AGENTS.md documents the new hard failure. Self-test 88 -> 96, each new rule with a negative control. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
let-fate-decide
A Claude Code skill that draws Tarot cards using secrets to inject
entropy into vague or underspecified planning decisions.
What It Does
When a prompt is sufficiently ambiguous, or the user explicitly invokes fate, this skill shuffles a 22-card Major Arcana deck and a 56-card Minor Arcana deck independently using cryptographic randomness, then deals the 12 Houses of the Zodiac spread. Each house receives one Major Arcana card followed by two Minor Arcana cards, and all 36 cards may appear reversed. Claude then interprets the spread and uses the reading to inform its approach.
Triggers
- Vague or ambiguous prompts where multiple reasonable approaches exist
- "I'm feeling lucky", "let fate decide", "dealer's choice", "surprise me", "whatever you think", "YOLO"
- Casual delegation ("whatever", "up to you", "your call", "idk", "just do something", "wing it", "I trust you", "doesn't matter", "do what you want", "I don't care", "any approach works", "you pick")
- Yu-Gi-Oh references ("heart of the cards", "I believe in the heart of the cards", "you've activated my trap card", "it's time to duel")
- Shrug-like brevity -- very short prompts that fully delegate the decision
- About to arbitrarily pick between 2+ valid approaches (draw cards instead)
- "Try again" on a system with no actual changes (redraw)
How It Works
The spread is drawn by the draw agent (agents/draw.md), dispatched as
let-fate-decide:draw.
- A Python script uses
secrets.randbelow()to perform Fisher-Yates shuffles - A Major Arcana deck and Minor Arcana deck are shuffled separately
- 12 houses are dealt, each with 1 Major Arcana and 2 Minor Arcana cards
- Each card has an independent 50% chance of being reversed
- Claude reads the house reference files and drawn cards' meaning files
- Claude interprets the spread
- The interpretation informs the planning direction
The default spread records a conservative unordered-card entropy budget
exceeding 100 bits: roughly log2(C(22,12)) bits from Major Arcana selection,
log2(C(56,24)) bits from Minor Arcana selection (assuming
secrets.randbelow() is cryptographically secure), and 36 bits from
independent card reversal choices. Exact values are computed at draw time and
reported in the JSON output under entropy_bits. The actual ordered assignment
of cards to houses contains more entropy.
In security, audit, and correctness workflows, this skill is only a discovery and hypothesis-generation aid. It can suggest where to look next, but evidence from review, testing, reproduction, or formal reasoning must decide whether a risk is real or resolved.
Reference Organization
Each zodiac house has its own markdown file under houses/. These files
explain the house's technical meaning for building new projects, vulnerability
discovery, correctness verification, and recurring workflows from the local
security and engineering workflows this skill supports.
references/TECHNICAL_CONTEXT_LENSES.md distills those workflow themes into
audit-stage, evidence-mode, domain, failure-class, and human/organizational
lenses.
Each of the 78 Tarot cards has its own markdown file:
cards/major/- 22 Major Arcana (The Fool through The World)cards/wands/- 14 Wands (Ace through King)cards/cups/- 14 Cups (Ace through King)cards/swords/- 14 Swords (Ace through King)cards/pentacles/- 14 Pentacles (Ace through King)