The audit family: evidence-grounded postmortems of finished work with a browsable index, a test-suite audit (does each green test still mean anything), a controls audit (census and live-fire over hooks, validators, and reminders), and the adversarial-verify primitive with its control-builder agent that the audits dispatch to
Plugin postmortem from marketplace fb-claude-skills in fblissjr/fb-claude-skills. Commands read from the repo's manifest — 10 days ago.
adversarial-verifyVerify something about to be trusted by trying to refute it, in two separate judgments: construct the experiment that would prove it false (dispatched to the control-builder agent), then independently verify the attempt actually reached the subject before believing either outcome. Use when the user
control-auditCensus and live-fire audit of a repo's controls: everything check-shaped that fires outside the test suite (git hooks, Claude Code hooks, CLI validators, ambient reminders). Per control, four slots are re-derived from current code — fires-via, guarded-by, retirement-condition, disclosed-uncontrolled
postmortem-indexBuild a browsable HTML index of every postmortem in a repo — date, mode, scope, one-line conclusion, and the artifacts each one examined, plus a by-artifact view answering 'has anything been written about this file'. Reads frontmatter only; superseded postmortems are shown and marked, never hidden.
postmortemEvidence-grounded postmortem of finished work — a session, a feature, or a span of sessions. Recovers what went well, what did not, deviations from plan, and test escapes; every finding must cite a concrete artifact, and empty sections are valid output. Use when the user says 'postmortem', 'run a po
test-auditAudit an existing test suite for meaning and drift: recover each test's motivating claim, verify its oracle can actually fail (spot mutation — deliberately break the guarded behavior, confirm red), and map the reachability envelope (conditions the harness never exercises). Classifies tests as load-b
control-builderTakes a claim and builds the control that would refute it, runs it, and reports which way it went. Delegate when something is about to be trusted - a new check, a threshold, a "this technique helps" belief, or a green result where you expected red. Reports the outcome; does not argue for it.
Structured, user-invoked standing prompts for long runs on any project: /design-scoreboard chooses what to climb and what must not get worse, /improve and /optimize climb it against the repo's AGENTS.md sections, and /trim-agents-md keeps that hub short. Each keeps state in files and reports what needs the owner first
4 skills · 2 agents · hooks
MECE decomposition for breaking work into mutually exclusive, collectively exhaustive parts. A skill you invoke when decomposing something -- it ships no SessionStart hook, because the method is needed at a decision point the model can recognise, not before the first action of every session.
5 skills · 1 MCP server
Skills for how the agent writes and explains: an American plain-language house style, voice matching, replacing prose with compact visuals, and re-pitching a message that did not land
4 skills · 1 command
Hand a task to Gemini and get a structured answer back -- video and audio analysis, perceptual image comparison, or any ad-hoc question with model, thinking level, and schema set per call. Explicit, auditable, one run directory per call.
1 skill · 2 commands · hooks
On-demand references for what a repo cannot state by being read: the house Python pinning policy and the two mistakes behind most Pydantic/Pyright diagnostic walls; documentation conventions (last-updated dates, where unshared notes and session logs live, how dependency changes are recorded, no decorative counts in prose); and a uv and bun dependency CVE audit, transitives included. No hooks or ambient injection: conventions belong in each repo's own always-loaded files.
3 skills
Tools for analyzing, polishing, and managing Claude Code plugins
1 skill · 2 agents
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.