agent coding skill risk: low

Codebase Evidence Wiki QA Responder

Answer repository questions using only source code evidence by searching files, reading content, and synthesizing responses with inline citations and a Key Files table.

SKILL 1 file

SKILL.md

Download

---
name: antigravity-awesome-skills-wiki-qa
description: "Answer repository questions grounded entirely in source code evidence. Use when user asks a question about the codebase, user wants to understand a specific file, function, or component, or user asks /\"how does X work/\" or /\"where is Y defined/\"."
---
# Wiki Q&A

Answer repository questions grounded entirely in source code evidence.

## When to Use
- User asks a question about the codebase
- User wants to understand a specific file, function, or component
- User asks "how does X work" or "where is Y defined"

## Procedure

1. Detect the language of the question; respond in the same language
2. Search the codebase for relevant files
3. Read those files to gather evidence
4. Synthesize an answer with inline citations

## Response Format

- Use `##` headings, code blocks with language tags, tables, bullet lists
- Cite sources inline: `(src/path/file.ts:42)`
- Include a "Key Files" table mapping files to their roles
- If information is insufficient, say so and suggest files to examine

## Rules

- ONLY use information from actual source files
- NEVER invent, guess, or use external knowledge
- Think step by step before answering

### When to Use
This skill is applicable to execute the workflow or actions described in the overview.

## Limitations
- Use this skill only when the task clearly matches the scope described above.
- Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
- Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.

REQUIRED CONTEXT

user question about the codebase

ROLES & RULES

ONLY use information from actual source files
NEVER invent, guess, or use external knowledge
Think step by step before answering

EXPECTED OUTPUT

Format

markdown

Schema

markdown_sections · ## headings, code blocks with language tags, tables, bullet lists, Key Files table, inline citations (src/path/file.ts:42)

Constraints

Use ## headings, code blocks with language tags, tables, bullet lists
Cite sources inline as (src/path/file.ts:42)
Include a Key Files table mapping files to their roles
If information is insufficient, say so and suggest files to examine
ONLY use information from actual source files
Think step by step before answering

SUCCESS CRITERIA

Answer repository questions grounded entirely in source code evidence
Detect the language of the question; respond in the same language
Synthesize an answer with inline citations
Include a Key Files table mapping files to their roles
If information is insufficient, say so and suggest files to examine

FAILURE MODES

May invent, guess, or use external knowledge
May omit required Key Files table or inline citations

CAVEATS

Dependencies

codebase access
source files

Missing context

Available tools or functions for searching/reading files
Target programming languages or repository scope

Ambiguities

Duplicate 'When to Use' section appears twice with conflicting content
References 'overview' which is not present in the prompt
Does not specify the mechanism or tools for 'Search the codebase' and 'Read those files'

QUALITY

OVERALL: 0.68
CLARITY: 0.70
SPECIFICITY: 0.60
REUSABILITY: 0.75
COMPLETENESS: 0.65

IMPROVEMENT SUGGESTIONS

Remove the duplicate 'When to Use' section and the sentence referencing a non-existent overview
Add concrete steps or tool-call examples under Procedure for searching and reading files
Specify output length expectations or example responses in Response Format

USAGE

Copy the prompt above and paste it into your AI of choice — Claude, ChatGPT, Gemini, or anywhere else you're working. Replace any placeholder sections with your own context, then ask for the output.