Agent-almanac analyze-codebase-workflow

install
source · Clone the upstream repo
git clone https://github.com/pjt222/agent-almanac
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/pjt222/agent-almanac "$T" && mkdir -p ~/.claude/skills && cp -r "$T/i18n/wenyan-ultra/skills/analyze-codebase-workflow" ~/.claude/skills/pjt222-agent-almanac-analyze-codebase-workflow-071741 && rm -rf "$T"
manifest: i18n/wenyan-ultra/skills/analyze-codebase-workflow/SKILL.md
source content

析庫流

掃庫自察數流、檔 I/O、依,出註計可手精。

  • 入新庫須知數流→用
  • 項無 PUT 註而起 putior→用
  • 文前審項數流→用
  • annotate-source-files
    前備註計→用

  • :庫或源處之路
  • :注定子處(默全庫)
  • :含/排之言(默諸察)
  • :察範:唯入、唯出、或兩(默兩+依)

一:掃庫構

識源檔與其言以知 putior 可析者。

library(putior)

# List all supported languages and their extensions
list_supported_languages()
list_supported_languages(detection_only = TRUE)  # Only languages with auto-detection

# Get supported extensions
exts <- get_supported_extensions()

以列檔知庫成:

# Count files by extension in the target directory
find /path/to/repo -type f | sed 's/.*\.//' | sort | uniq -c | sort -rn | head -20

得:庫中諸延附數。映於

get_supported_extensions()
知覆。

敗:庫無檔合支延→putior 不能自察。或言支但檔用非標延。

二:察言覆

各察言驗自察紋有否。

# Check which languages have auto-detection patterns (18 languages, 902 patterns)
detection_langs <- list_supported_languages(detection_only = TRUE)
cat("Languages with auto-detection:\n")
print(detection_langs)

# Get pattern counts for specific languages found in the repo
for (lang in c("r", "python", "javascript", "sql", "dockerfile", "makefile")) {
  patterns <- get_detection_patterns(lang)
  cat(sprintf("%s: %d input, %d output, %d dependency patterns\n",
    lang,
    length(patterns$input),
    length(patterns$output),
    length(patterns$dependency)
  ))
}

得:各言紋數印。R 124、Python 159、JavaScript 71 等。

敗:言返無紋→支手註而非自察。計手註此檔。

三:行自察

於標處行

put_auto()
以發流元。

# Full auto-detection
workflow <- put_auto("./src/",
  detect_inputs = TRUE,
  detect_outputs = TRUE,
  detect_dependencies = TRUE
)

# Exclude build scripts and test helpers from scanning
workflow <- put_auto("./src/",
  detect_inputs = TRUE,
  detect_outputs = TRUE,
  detect_dependencies = TRUE,
  exclude = c("build-", "test_helper")
)

# View detected workflow nodes
print(workflow)

# Check node count
cat(sprintf("Detected %d workflow nodes\n", nrow(workflow)))

大庫漸析子處:

# Analyze specific subdirectories
etl_workflow <- put_auto("./src/etl/")
api_workflow <- put_auto("./src/api/")

得:數框含

id
label
input
output
source_file
等欄。各行為一察流步。

敗:果空→源或無可識 I/O 紋。試開除錯:

workflow <- put_auto("./src/", log_level = "DEBUG")
察掃檔與配紋。

四:生初圖

繪自察流以評覆與識缺。

# Generate diagram from auto-detected workflow
cat(put_diagram(workflow, theme = "github"))

# With source file info for traceability
cat(put_diagram(workflow, show_source_info = TRUE))

# Save to file for review
writeLines(put_diagram(workflow, theme = "github"), "workflow-auto.md")

得:Mermaid 流圖示察點以數流邊接。點標當有意函/檔名。

敗:圖示斷點→自察找 I/O 紋而不能推接。常事——接由出檔名配入檔名而生。註計(次步)解缺。

五:出註計

生構文示所發及待手註者。

# Generate annotation suggestions
put_generate("./src/", style = "single")

# For multiline style (more readable for complex workflows)
put_generate("./src/", style = "multiline")

# Copy suggestions to clipboard for easy pasting
put_generate("./src/", output = "clipboard")

文以覆評:

## Annotation Plan

### Auto-Detected (no manual work needed)
- `src/etl/extract.R` — 3 inputs, 2 outputs detected
- `src/etl/transform.py` — 1 input, 1 output detected

### Needs Manual Annotation
- `src/api/handler.js` — Language supported but no I/O patterns matched
- `src/config/setup.sh` — Only 12 shell patterns; complex logic missed

### Not Supported
- `src/legacy/process.f90` — Fortran not in detection languages

### Recommended Connections
- extract.R output `data.csv` → transform.py input `data.csv` (auto-linked)
- transform.py output `clean.parquet` → load.R input (needs annotation)

得:清計分自察檔與待手註者,附各檔具體薦。

敗:

put_generate()
無出→確路正且含支言之源檔。

  • put_auto()
    於標處行而無誤
  • 察流至少一點(除非庫無可識 I/O)
  • put_diagram()
    自察流出有效 Mermaid
  • put_generate()
    為察紋之檔出註薦
  • 註計文已建附覆評

  • 掃過廣:庫根行
    put_auto(".")
    或含
    node_modules/
    .git/
    venv/
    等。標定源處
  • 盼全覆:自察找檔 I/O 與庫呼非商邏。常 40-60% 覆;餘待手註
  • 忽依
    detect_dependencies = TRUE
    旗捕
    source()
    import
    require()
    連腳。停則失跨檔接
  • 言不配:非標延(如
    .R
    .r
    .jsx
    .js
    )或不察。用
    get_comment_prefix()
    察延識否。
    Dockerfile
    Makefile
    等無延檔以全名配支
  • 大庫:百源檔以上→按模/處析以保圖可讀

  • install-putior
    — 前置:先裝 putior
  • annotate-source-files
    — 次步:按計加手註
  • generate-workflow-diagram
    — 註畢生末圖
  • configure-putior-mcp
    — 用 MCP 工為互析