Fidelity Reviewer — Development agent for Claude Code
lit-panel 文学评审团第 01 席(忠实审读).
How to install Fidelity Reviewer
Installs to ~/.claude/agents/rockiez-lit-panel-fidelity-reviewer.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/Rockiez/lit-panel/HEAD/agents/fidelity-reviewer.md -o ~/.claude/agents/rockiez-lit-panel-fidelity-reviewer.md Restart Claude Code, or start a new session, for it to be picked up.
What Fidelity Reviewer does
name: lit-fidelity description: lit-panel 文学评审团第 01 席(忠实审读)。仅由 lit-panel 编排流程(SKILL.md//lit-review//lit-compare)调度,当调用方提供 `--source` 来源素材时激活。只认来源:对文本中的具体断言做五态标注(SUPPORTED/PERMISSIBLE_INFERENCE/UNSUPPORTED/CONTRADICTED/UNVERIFIABLE),不评价文笔、结构或情感。拥有红线判定权。 tools: Read, Grep, Glob
你是文学评审团(lit-panel)第 01 席评审员,agent name `lit-fidelity`——忠实审读席。
你是谁与自在观点
你只认一件事:文本里的具体断言,能不能在 `--source` 来源素材里找到支持依据。你的参照系是来源文本本身,不是"这段读起来合不合理""这段写得好不好"。
你只关心:姓名/称谓/排行/亲属关系、年龄/日期/数量/时间跨度、人物的知识与技能归属、编辑任务或来源明确要求保
Alternatives in Development
- Retool Fidelity Review — The fourth review for a Retool migration slice — walks the export query by query and asks whether each behavio 7.2k ★
- Writer Critic — Manuscript critic that reviews paper drafts for structure, claims-evidence alignment, identification fidelity 250 ★
- Part 16: autoDream — Automatic Memory Consolidation — Note: OpenClaw 2026.4+ has a built-in dreaming system (Part 22) inside memory-core 232 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Brief Adherence Reviewer
lit-panel 文学评审团第 10 席(编辑意图审读)。仅由 lit-panel 编排流程(SKILL.md//lit-review//lit-compare)调度,当调用方提供 --brief 时激活。审 brie
Character Reviewer
lit-panel 文学评审团第 05 席(人物与心理审读)。仅由 lit-panel 编排流程(SKILL.md//lit-review//lit-compare)调度,总是激活。审动机连续性、对话口吻辨识度、心理描写
Verify Findings
Final verification pass over a set of findings before they are reported. Deduplicates, removes anything contra
Uxaudit L3 Judge
L3-vision Judge for the uxaudit pipeline. Reads ONE captured screenshot plus a check-specific prompt.md (and o
Uxaudit L4 Judge
L4-journey Judge for the uxaudit pipeline. Reads the per-journey capture directory (screenshots + steps.json +
Evidence Validator
Single-purpose verification agent — given a finding (a claim + file:line reference), reads the actual source a
Related Skills
Vt Compliance
VT compliance testing -- run escape sequence tests against teru's VtParser, validate CSI/SGR/OSC/DCS handling,
Aiproxer
Connect any LLM-powered client app, such as a coding agent, to any supported inference backend/model.
Lit Compare
对两篇同源同任务的中文回忆录/叙事文本发起十一席文学评审团互盲对比评审(换序双判,禁总比分)