Field Report
- 5 repo stars
- Updated July 9, 2026
- nicholls-inc/claude-code-marketplace
Generate structured performance reports on plugins, skills, and agents by analyzing Claude Code session conversations.
About
field-report analyzes Claude Code session conversations to produce structured performance reports on plugins, skills, and agents. A developer uses it to measure how their tooling actually performs in real sessions. It turns raw session logs into an evaluation of what is working.
- Structured performance reports
- Analyzes session conversations
- Covers plugins, skills, agents
Field Report by the numbers
- Data as of Jul 25, 2026 (Skillselion catalog sync)
/plugin marketplace add nicholls-inc/claude-code-marketplace/plugin install field-report@nichollsAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| repo stars | ★ 5 |
|---|---|
| Last updated | July 9, 2026 |
| Repository | nicholls-inc/claude-code-marketplace ↗ |
What it does
Generate structured performance reports on plugins, skills, and agents by analyzing Claude Code session conversations.
README.md
field-report — Claude Code Plugin
Generate structured performance reports on plugins, skills, and agents by analysing Claude Code session conversations. This plugin produces evidence-based narrative reports with actionable recommendations for artifact developers.
Installation
Point Claude Code at this plugin directory or install from the marketplace:
claude plugin install field-report@nicholls
Usage
/field-report
Generate a structured field report for one subject using one Claude Code session as evidence.
Usage: /field-report <session-id or description> <subject-name> [path/to/definition]
Examples:
/field-report ses_abc123 reason— Explicit session ID and subject/field-report "last session" crosscheck/reason— Natural language session selector/field-report "today's session" byfuglien crosscheck/agents/byfuglien.md— With explicit definition path
Analysis Dimensions
The report evaluates the subject across six key dimensions:
- Task Completion: Did the subject help complete the user goal?
- Instruction Adherence: How closely did execution match subject instructions?
- Tool Usage Patterns: Evaluation of tool selection and execution behavior.
- Conversation Efficiency: Measurement of progress quality and momentum.
- Error Handling: Inspection of observable failures and recoveries.
- Input Clarity: Evaluation of initial request clarity and clarification needs.
Report Output
Reports are saved to the field-reports/ directory in your working directory. Filenames follow the convention {subject-slug}--{session-id-short}--{YYYY-MM-DD}.md. Each report includes a header metadata block, detailed analysis sections, lessons learned, and a summary.
Sanitisation
Reports are privacy-safe. The plugin applies explicit redaction rules:
- STRIP: Email addresses, API keys, secrets, environment variable values, and personal names.
- KEEP: File paths, tool names, error messages, and code patterns.
Limitations
- Session discovery is heuristic. Explicit session IDs are more reliable than natural language descriptions.
- Each report covers exactly one session and one subject.
- Tool usage analysis depends on the session data format and availability of tool-call traces.
- No baseline comparisons or historical trending across multiple reports.
- Sessions must have at least 10 messages involving the subject to ensure sufficient evidence density.
Related skills
Grill MeHave an agent relentlessly pressure-test their plan or design until every decision branch is resolved and shared understanding is reached.691k196kGrill With DocsPressure-test architecture decisions against their live codebase and living documentation before writing production code.585k196k
Lark CalendarLet their coding agents read, create, and manage events inside Lark (Feishu) Calendar directly from natural language instructions.493kLark ImLet their coding agent send instant notifications, post updates, and receive commands directly inside Lark (Feishu) workspaces.493kLark ContactLet their AI coding agents read and write contacts, send messages, and manage user data inside Lark (Feishu).493kLark AttendanceAutomatically fetch and report daily attendance data from Lark (Feishu) into their own tools or agents.493k