Claude Code プラグイン
公開されている Claude Code marketplace からの 6 件のプラグインです。それぞれにスキル、スラッシュコマンド、エージェント、フック、MCP サーバーがまとめてインストールされます。
これらのフィルターに一致するプラグインを 6 件表示しています。 フィルターをクリア
Answer questions about past agent runs from the recording, not from memory. Three read-only stdlib scripts over the session transcripts Claude Code and Codex already write: a finder that locates the run by repo, time window, touched file or command text; a timeline that numbers every step with its exit code, lists every file changed (flagging changes the agent never mentioned) and walks back from any step with each causal edge labelled recorded or inferred by a named rule; and a claim checker that tests a summary's claims (ran, passed, edited, not-edited, read) against the run and catches tests that passed before a later edit and exit codes that belong to the last command of a pipe. Secrets are always redacted. A UserPromptSubmit hook gives the agent this session's transcript path when a prompt asks about past work. Ships cs-run-forensics agent, /cs-run-forensics and /cs-check-claims commands and 3 cited references.
A four-tier memory ladder for Claude Code where promotion is earned by recurrence, not asserted by confidence. L0 raw transcripts are never injected; L1 candidate atoms are gitignored and recalled on lexical relevance per prompt; L2 project context loads each session start; L3 stable persona is always loaded. L1 to L2 needs three distinct sessions spanning two distinct calendar days (a stated claim needs two; a verified claim is the only single-observation path); L2 to L3 needs two projects and thirty days. Two gates refuse rather than guess: a claim altered by the redaction pass never promotes on evidence alone, and a contradicted claim freezes until a human resolves it. Three hooks run the loop unattended and every one fails open. Nothing reaches a committed CLAUDE.md without an explicit human adopt. Ships 5 stdlib scripts, 3 hooks, a cs-memory-curator agent, /cs-memory, a JSON schema, and a 69-check validator. Concept from TencentCloud/TencentDB-Agent-Memory (MIT); no upstream code
Grade an agent setup from real conversation history — a rebuild of warpdotdev/common-skills' skill-doctor (MIT). Harvests recent local Claude Code and Codex sessions scoped to one repo, condenses and secret-redacts transcripts (12-pattern redactor, chmod-0600 artifacts, nothing uploaded), has the agent judge each transcript against two verbatim-preserved rubrics (efficiency, code quality — labels only), then runs a deterministic aggregation gate that derives every number from the label tables, refuses scores for unsampled sessions and suggestions that cite no scored session, and renders one self-contained zero-JS HTML report (dark-mode, print-to-PDF). Proposed skill edits stay staged as diffs until an explicit per-skill yes; zero suggestions is a valid, reportable success. Ships cs-skill-retro agent, /cs-skill-retro command, 3 stdlib tools, 3 references citing 7 sources each.
Turn screen recordings into ready-to-file bug drafts, specs, backlogs and meeting actions — local Whisper transcript + keyframes + OCR + wall-clock evidence via the talkthrough MCP server.
Memory-defragment/cleanup engine for agent memory, TWO lanes: class-B (memory + governance files) gets the semantic wash — a free mechanical Quick pass plus a code-enforced fidelity gate blocking any STRUCTURED-token drop by diff (links, dates, versions, frontmatter; prose fidelity is the paid semantic layer's job), consent-gated Full pass, every cut snapshot-backed (whole-run rollback); class-A (transcripts — the estate) gets byte-identity ULTRA archiving, never semantic-edited. Honest frame: slows how fast memory-overhead grows — it does not eliminate memory cost. Zero-dependency, offline, no API keys.
A local YouTube memory: fetches transcripts (captions, Whisper fallback), keeps them searchable on your machine, and writes summaries, timestamped highlights, Q&A, decks and daily digests. Ships the yt + yt-agent skills and /yt-* commands; no API key.