Contributor Analyzer
Unified skill for engineering contributor analysis. Combines annual review generation, historical progress tracking, production ownership mapping, and qualitative judgment — all driven by reading actual commit diffs.
Skill Format
Each reference file covers one analysis concern:
- Quick Reference: Summary tables and key thresholds
- Process Steps: Bash commands and orchestration
- Templates: Output formats for reports and profiles
- Deep Dive: Full context with examples from real-world 1,091-commit analysis
Impact ratings: CRITICAL (blocks accuracy), HIGH (major insight), MEDIUM (enriches analysis)
When to Apply
Reference these guidelines when:
- Running annual or quarterly performance reviews
- Assessing promotion readiness with evidence
- Tracking contributor progress over time (incremental updates)
- Mapping production code ownership and risk
- Comparing multiple engineers for team decisions
- Identifying single points of failure in codebase ownership
Priority-Ordered References
| Priority |
Reference |
Impact |
Description |
| 1 |
agent-context-management.md |
CRITICAL |
Batch sizing, file-based output, context budget |
| 2 |
annual-review-process.md |
CRITICAL |
7-phase analysis: identity, metrics, diffs, bugs, quality, report, comparison |
| 3 |
accuracy-analysis.md |
HIGH |
Bug introduction detection, accuracy rate calculation, benchmarks |
| 4 |
code-quality-catalog.md |
HIGH |
Anti-pattern catalog (9 patterns) and strength catalog (8 patterns) |
| 5 |
qualitative-judgment.md |
HIGH |
Engineering wisdom, situational decisions, growth trajectory |
| 6 |
report-templates.md |
HIGH |
Report sections, rating scale, promotion framework, comparison format |
| 7 |
production-ownership.md |
MEDIUM |
Production branch ownership, domain map, SPOF detection |
| 8 |
historical-progress.md |
MEDIUM |
Incremental profiles, plateau detection, period-over-period tracking |
Quick Reference
Critical: Context Management
Before launching ANY analysis, count commits:
git log --author="EMAIL" --after="YEAR-01-01" --before="YEAR+1-01-01" --oneline | wc -l
Batch sizing (hard limits from real failures):
| Commits |
Action |
| 1-40 |
Read in main session |
| 41-70 |
Single sub-task, writes findings to file |
| 71-90 |
Split into 2 sub-tasks |
| 91+ |
WILL FAIL — split into 3+ or monthly sub-tasks |
Sub-tasks write to files, return 3-line summaries. Never return raw analysis inline.
Critical: Annual Review Process
7 phases, sequential:
- Identity Discovery — find all git email variants
- Metrics — commits, PRs, reviews, lines (git + gh CLI)
- Read ALL Diffs — quarterly parallel sub-tasks, file-based output
- Bug Introduction — self-reverts, crash-fixes, same-day fixes, hook bypass
- Code Quality — anti-patterns and strengths from diff reading
- Report Generation — structured markdown with rating + promotion assessment
- Comparison — multi-engineer ranking with evidence
High: Accuracy Rate
Effective Accuracy = 100% - (fix-related commits / total commits)
| Rate |
Assessment |
| >90% |
Excellent |
| 85-90% |
Good |
| 80-85% |
Concerning |
| <80% |
Poor — significant rework |
High: Tool Separation
- Local
git: ALL commit-level analysis (log, show, diff, blame, shortlog)
gh CLI: ONLY for PR counts, review counts, user verification- Never use
gh to read diffs — local git is faster with no rate limits
References
Full documentation with process steps and templates in references/:
Analysis Process
| File |
Impact |
Description |
annual-review-process.md |
CRITICAL |
Complete 7-phase review process with bash commands |
agent-context-management.md |
CRITICAL |
Batch sizing, file-based output, multi-engineer orchestration |
accuracy-analysis.md |
HIGH |
Bug detection commands, accuracy formula, benchmarks |
code-quality-catalog.md |
HIGH |
9 anti-patterns + 8 strengths with severity/significance |
Judgment & Assessment
| File |
Impact |
Description |
qualitative-judgment.md |
HIGH |
Engineering wisdom indicators, growth trajectory, promotion signals |
report-templates.md |
HIGH |
Required report sections, rating scale, comparison format |
Ownership & Progress
| File |
Impact |
Description |
production-ownership.md |
MEDIUM |
Production file identification, domain map, SPOF detection |
historical-progress.md |
MEDIUM |
Incremental profiles, plateau detection, cumulative profiles |
Searching References
# Find by keyword
grep -rl "accuracy" references/
grep -rl "promotion" references/
grep -rl "ownership" references/
grep -rl "anti-pattern" references/
Problem to Reference Mapping
| Problem |
Start With |
| Annual review for 1 engineer |
annual-review-process.md then report-templates.md |
| Comparing 2+ engineers for promotion |
annual-review-process.md then qualitative-judgment.md |
| Engineer has 200+ commits |
agent-context-management.md (read FIRST) |
| Session keeps running out of context |
agent-context-management.md |
| Is this engineer ready for promotion? |
qualitative-judgment.md then accuracy-analysis.md |
| Who owns the payment system? |
production-ownership.md |
| Track progress since last review |
historical-progress.md |
| Quality assessment from code |
code-quality-catalog.md then accuracy-analysis.md |
| Single points of failure in team |
production-ownership.md |
| Plateau detection |
historical-progress.md |
Full Compiled Document
For the complete guide with all references expanded: AGENTS.md
Usage Examples
# Annual review (single engineer)
Analyze @alice for 2025 annual review in repo org/repo.
Git email: alice@company.com. Write to ./alice-2025-review.md
# Multi-engineer comparison with promotion decision
Analyze @alice, @bob, @charlie for 2025 reviews.
I need to decide which 2 get promoted.
# Production ownership mapping
Analyze production code ownership in this repo.
# Incremental progress update
Update @alice's progress profile with latest contributions.
# Critical path risk detection
Find single points of failure in production code ownership.
# Plateau detection
Has @bob plateaued? Check last 3 quarters.
1---2name: contributor-analyzer3description: Analyzes engineering contributors for annual reviews, progress tracking, and production ownership. Reads every commit diff (not sampling), calculates accuracy rates, assesses engineering maturity through qualitative judgment, and generates promotion-ready performance reviews. Works with any git repository using local git for commit analysis and gh CLI for GitHub PR/review metadata.4license: MIT5---6
7# Contributor Analyzer
8
9Unified skill for engineering contributor analysis. Combines annual review generation, historical progress tracking, production ownership mapping, and qualitative judgment — all driven by reading actual commit diffs.
10
11## Skill Format
12
13Each reference file covers one analysis concern:
14
15- **Quick Reference**: Summary tables and key thresholds
16- **Process Steps**: Bash commands and orchestration
17- **Templates**: Output formats for reports and profiles
18- **Deep Dive**: Full context with examples from real-world 1,091-commit analysis
19
20**Impact ratings**: CRITICAL (blocks accuracy), HIGH (major insight), MEDIUM (enriches analysis)
21
22## When to Apply
23
24Reference these guidelines when:
25- Running annual or quarterly performance reviews
26- Assessing promotion readiness with evidence
27- Tracking contributor progress over time (incremental updates)
28- Mapping production code ownership and risk
29- Comparing multiple engineers for team decisions
30- Identifying single points of failure in codebase ownership
31
32## Priority-Ordered References
33
34| Priority | Reference | Impact | Description |
35|----------|-----------|--------|-------------|
36| 1 | `agent-context-management.md` | CRITICAL | Batch sizing, file-based output, context budget |
37| 2 | `annual-review-process.md` | CRITICAL | 7-phase analysis: identity, metrics, diffs, bugs, quality, report, comparison |
38| 3 | `accuracy-analysis.md` | HIGH | Bug introduction detection, accuracy rate calculation, benchmarks |
39| 4 | `code-quality-catalog.md` | HIGH | Anti-pattern catalog (9 patterns) and strength catalog (8 patterns) |
40| 5 | `qualitative-judgment.md` | HIGH | Engineering wisdom, situational decisions, growth trajectory |
41| 6 | `report-templates.md` | HIGH | Report sections, rating scale, promotion framework, comparison format |
42| 7 | `production-ownership.md` | MEDIUM | Production branch ownership, domain map, SPOF detection |
43| 8 | `historical-progress.md` | MEDIUM | Incremental profiles, plateau detection, period-over-period tracking |
44
45## Quick Reference
46
47### Critical: Context Management
48
49**Before launching ANY analysis, count commits:**
50```bash
51git log --author="EMAIL" --after="YEAR-01-01" --before="YEAR+1-01-01" --oneline | wc -l
52```
53
54**Batch sizing (hard limits from real failures):**
55
56| Commits | Action |
57|---------|--------|
58| 1-40 | Read in main session |
59| 41-70 | Single sub-task, writes findings to file |
60| 71-90 | Split into 2 sub-tasks |
61| 91+ | WILL FAIL — split into 3+ or monthly sub-tasks |
62
63**Sub-tasks write to files, return 3-line summaries.** Never return raw analysis inline.
64
65### Critical: Annual Review Process
66
677 phases, sequential:
681. **Identity Discovery** — find all git email variants
692. **Metrics** — commits, PRs, reviews, lines (git + gh CLI)
703. **Read ALL Diffs** — quarterly parallel sub-tasks, file-based output
714. **Bug Introduction** — self-reverts, crash-fixes, same-day fixes, hook bypass
725. **Code Quality** — anti-patterns and strengths from diff reading
736. **Report Generation** — structured markdown with rating + promotion assessment
747. **Comparison** — multi-engineer ranking with evidence
75
76### High: Accuracy Rate
77
78```
79Effective Accuracy = 100% - (fix-related commits / total commits)
80```
81
82| Rate | Assessment |
83|------|-----------|
84| >90% | Excellent |
85| 85-90% | Good |
86| 80-85% | Concerning |
87| <80% | Poor — significant rework |
88
89### High: Tool Separation
90
91- **Local `git`**: ALL commit-level analysis (log, show, diff, blame, shortlog)
92- **`gh` CLI**: ONLY for PR counts, review counts, user verification
93- **Never use `gh` to read diffs** — local git is faster with no rate limits
94
95## References
96
97Full documentation with process steps and templates in `references/`:
98
99### Analysis Process
100
101| File | Impact | Description |
102|------|--------|-------------|
103| `annual-review-process.md` | CRITICAL | Complete 7-phase review process with bash commands |
104| `agent-context-management.md` | CRITICAL | Batch sizing, file-based output, multi-engineer orchestration |
105| `accuracy-analysis.md` | HIGH | Bug detection commands, accuracy formula, benchmarks |
106| `code-quality-catalog.md` | HIGH | 9 anti-patterns + 8 strengths with severity/significance |
107
108### Judgment & Assessment
109
110| File | Impact | Description |
111|------|--------|-------------|
112| `qualitative-judgment.md` | HIGH | Engineering wisdom indicators, growth trajectory, promotion signals |
113| `report-templates.md` | HIGH | Required report sections, rating scale, comparison format |
114
115### Ownership & Progress
116
117| File | Impact | Description |
118|------|--------|-------------|
119| `production-ownership.md` | MEDIUM | Production file identification, domain map, SPOF detection |
120| `historical-progress.md` | MEDIUM | Incremental profiles, plateau detection, cumulative profiles |
121
122## Searching References
123
124```bash
125# Find by keyword
126grep -rl "accuracy" references/
127grep -rl "promotion" references/
128grep -rl "ownership" references/
129grep -rl "anti-pattern" references/
130```
131
132## Problem to Reference Mapping
133
134| Problem | Start With |
135|---------|------------|
136| Annual review for 1 engineer | `annual-review-process.md` then `report-templates.md` |
137| Comparing 2+ engineers for promotion | `annual-review-process.md` then `qualitative-judgment.md` |
138| Engineer has 200+ commits | `agent-context-management.md` (read FIRST) |
139| Session keeps running out of context | `agent-context-management.md` |
140| Is this engineer ready for promotion? | `qualitative-judgment.md` then `accuracy-analysis.md` |
141| Who owns the payment system? | `production-ownership.md` |
142| Track progress since last review | `historical-progress.md` |
143| Quality assessment from code | `code-quality-catalog.md` then `accuracy-analysis.md` |
144| Single points of failure in team | `production-ownership.md` |
145| Plateau detection | `historical-progress.md` |
146
147## Full Compiled Document
148
149For the complete guide with all references expanded: `AGENTS.md`
150
151## Usage Examples
152
153```
154# Annual review (single engineer)
155Analyze @alice for 2025 annual review in repo org/repo.
156Git email: alice@company.com. Write to ./alice-2025-review.md
157
158# Multi-engineer comparison with promotion decision
159Analyze @alice, @bob, @charlie for 2025 reviews.
160I need to decide which 2 get promoted.
161
162# Production ownership mapping
163Analyze production code ownership in this repo.
164
165# Incremental progress update
166Update @alice's progress profile with latest contributions.
167
168# Critical path risk detection
169Find single points of failure in production code ownership.
170
171# Plateau detection
172Has @bob plateaued? Check last 3 quarters.
173```
majiayu000