Performance

Codex compatibility note:

>

- Invoke repository skills with $skill-name in Codex; this mirrored copy rewrites legacy Claude /skill-name references.

- Task tracker mandate: BEFORE executing any workflow or skill step, create/update task tracking for all steps and keep it synchronized as progress changes.

- User-question prompts mean to ask the user directly in Codex.

- Ignore Claude-specific mode-switch instructions when they appear.

- Strict execution contract: when a user explicitly invokes a skill, execute that skill protocol as written.

- Subagent authorization: when a skill is user-invoked or AI-detected and its protocol requires subagents, that skill activation authorizes use of the required spawn_agent subagent(s) for that task.

- Do not skip, reorder, or merge protocol steps unless the user explicitly approves the deviation first.

- For workflow skills, execute each listed child-skill step explicitly and report step-by-step evidence.

- If a required step/tool cannot run in this environment, stop and ask the user before adapting.

<!-- CODEX:PROJECT-REFERENCE-LOADING:START -->

Codex Project-Reference Loading (No Hooks)

Codex uses static project-reference loading instead of runtime-injected project docs. When coding, planning, debugging, testing, or reviewing, open project docs explicitly using this routing.

Always read:

• docs/project-config.json (project-specific paths, commands, modules, and workflow/test settings)
• docs/project-reference/docs-index-reference.md (routes to the full docs/project-reference/* catalog)
• docs/project-reference/lessons.md (always-on guardrails and anti-patterns)

Missing/stale context route: If docs/project-config.json, the docs index, lessons.md, CLAUDE.md, AGENTS.md, or any task-required reference doc is missing or stale, auto-run $project-init or the narrow setup route ($project-config, $docs-init, $scan-all, $scan --target=<key>, $claude-md-init) before ordinary project-specific work. If Codex mirrors or AGENTS.md are missing/stale, ask the user to run $sync-codex; do not auto-run it.

Situation-based docs:

• Backend/CQRS/API/domain/entity changes: backend-patterns-reference.md, domain-entities-reference.md, project-structure-reference.md
• Frontend/UI/styling/design-system: frontend-patterns-reference.md, scss-styling-guide.md, design-system/README.md
• Spec authoring, docs/specs/ pathing, or TC format: feature-spec-reference.md, spec-system-reference.md, spec-principles.md
• Behavior/public-contract changes or spec-test-code sync: workflow-spec-test-code-cycle-reference.md plus the spec docs above
• Derived spec indexes/ERDs/reimplementation guides: spec-system-reference.md and source Feature Specs under docs/specs/
• Integration test implementation/review: integration-test-reference.md
• E2E test implementation/review: e2e-test-reference.md
• Code review/audit work: code-review-rules.md plus domain docs above based on changed files

Do not read all docs blindly. Start from docs-index-reference.md, then open only relevant files for the task.

<!-- CODEX:PROJECT-REFERENCE-LOADING:END -->

Quick Summary

Goal: Plan and generate a developer KPI-style quality-work report from local git history only. Workflow:

1. Set Goal + Plan - Declare the goal, trigger $plan, and create tasks per contributor. 2. Collect Packets - Run scripts/git-developer-performance.cjs to create commit inventory and work packets. 3. Analyze Work - Read patches per contributor; estimate value, story points, man-days, and quality impact. 4. Synthesize Report - Write quality-work-summary.md and evidence-proof.md outside .claude. Key Rules:

• Use only local git history. Do not query external services.
• Consolidate people by identity map, then normalized email, then high-confidence aliases such as DOMAIN\first.lastpart matching a full name; use --identity-map for exceptions.
• This is a large task: plan first, then create one todo task per contributor.
• The script collects evidence; AI must read changes and synthesize contributed value.
• Treat KPI values as evidence-based estimates, not a complete HR assessment.
• Report both man_days_traditional (no AI) and man_days_ai (AI coding assistant with project context).
• Traverse full merged branch history (not only first-parent) and attribute shared feature-branch implementation to each developer's own direct commits; merge authors get integration/admin signal unless conflict-resolution changes are explicitly inspected.
• Estimate implementation SP from direct authored diffs first; zero-change merge/admin commits are integration signal only.
• Discount generated files, migration designers, docs/spec output, i18n sorting, lockfiles, and repeated follow-up churn.
• For velocity mismatch or recheck requests, synthesize each contributor's direct authored work as one "giant commit" first, then split into atomic 1/2/3/5/8/13 SP clusters.
• Persist large rechecks to a report file outside .claude before finalizing, so context loss cannot erase evidence.
• Separate product/domain delivery, infrastructure/tooling work, docs/generated churn, and merge/admin integration; do not mix them silently into one velocity number.
• Run a velocity sanity check: both man-day ranges must be plausible for active days and the selected period.
• Keep output outside .claude; default root is reports/developer-performance/.

Git Developer Performance

Use when the user asks for developer KPI/performance, productivity, contribution value, story-point estimates, man-day estimates, quality impact, or quality-work reporting from git commits.

Required AI Workflow

Before analysis, set or declare this goal:

Plan and generate a developer performance quality-work report from local git history, then execute the plan and produce the report.

Then trigger $plan or create equivalent plan artifacts. This skill is not a commit-list export. It requires reading direct commits and merge/admin commits per contributor, then synthesizing value. Use ultrathink/deep analysis for final synthesis when contributor count or churn is high.

Command

node .claude/skills/git-developer-performance/scripts/git-developer-performance.cjs [options]

Options: --branch <ref> defaults to develop then main; --days <n> defaults to 60; --since <date> overrides days; --until <date> defaults now; --out <dir> defaults to reports/developer-performance; --identity-map <csv> accepts identity,email,displayName,id; --json prints machine-readable result.

Examples:

node .claude/skills/git-developer-performance/scripts/git-developer-performance.cjs
node .claude/skills/git-developer-performance/scripts/git-developer-performance.cjs --branch release/1.4 --days 30
node .claude/skills/git-developer-performance/scripts/git-developer-performance.cjs --since 2026-01-01 --until 2026-03-31 --out reports/dev-performance-q1

Output

Creates a timestamped run folder containing:

• summary.md - team evidence report, authored signal sort, warnings, and integration/admin activity.
• analysis-plan.md - AI execution plan with one task per contributor.
• work-packets/*.md - per-contributor commit/change packets for qualitative analysis.
• quality-work-summary.md and evidence-proof.md - AI-written value synthesis and proof appendix.
• analysis/ - target folder for AI-written per-contributor synthesis.
• contributors.csv, commits.csv, developers/*.md, data/*.json - source evidence and deterministic aggregates.

Analysis Rules

• Read references/analysis-workflow.md before final synthesis.
• Treat contributors as people consolidated by identity map/email/high-confidence aliases, not raw display names.
• Count distinct contributors, then create one todo task per contributor from analysis-plan.md.
• For each contributor, inspect direct authored commits and merge/admin commits from work-packets/*.md.
• Use git show --stat --find-renames <hash> and targeted patches for high-impact commits.
• When several developers contribute to one feature branch, analyze each contributor's direct commits separately and never give the whole feature's implementation SP to the merge author or PR owner.
• Estimate work clusters with 1/2/3/5/8/13 story points, no-AI man-days, and AI-assisted man-days; state confidence.
• If a displayed theme is more than 13 SP, state that it is a sum of smaller atomic clusters, not one unsplit story.
• Do not add implementation SP for zero-file merge/admin commits; mention them separately as integration/admin signal.
• Discount non-implementation churn before estimating: generated code, EF designer snapshots, docs/specs, i18n sorting, lockfiles, and repeated follow-ups.
• Reconcile final SP/man-day totals against authored active days and team velocity intuition; if implausible, re-audit before delivery.
• Analyze contributed value: features/changes, bug fixes, refactors, tests/docs, integration/admin, and code quality.
• If there are many contributors, split contributor tasks across subagents with disjoint developer lists.
• Review identity and bulk-change warnings before comparing contributors.
• State that report quality depends on local git data quality when history is incomplete, stale, squashed, or bot/shared authors exist.

Verification

Before delivering a generated report:

1. Run node --test .claude/skills/git-developer-performance/tests/*.test.cjs. 2. Run the command for the requested repo/range. 3. Confirm the output path is outside .claude.

<!-- SYNC:ai-mistake-prevention -->

AI Mistake Prevention — Failure modes to avoid on every task:

>

Re-read files after context changes. Context compaction, resume, or long-running work can make memory stale; verify current files before acting.

Verify generated content against source evidence. AI hallucinates APIs, names, claims, and document facts. Check the relevant source before documenting or referencing.

Check downstream references before deleting or renaming. Removing an artifact can stale docs, generated mirrors, configs, and callers; map references first.

Trace the full impact chain after edits. Changing a definition can miss derived outputs and consumers. Follow the affected chain before declaring done.

Verify ALL affected outputs, not just the first. One green check is not all green checks; validate every output surface the change can affect.

Assume existing values are intentional — ask WHY before changing. Before changing a constant, limit, flag, wording, or pattern, read nearby context and history.

Surface ambiguity before acting — don't pick silently. Multiple valid interpretations require an explicit question or stated assumption with risk.

Keep shared guidance role-relevant. Universal guidance must help every receiving skill or agent; code-specific obligations belong only in code-specific protocols.

<!-- /SYNC:ai-mistake-prevention -->

<!-- SYNC:critical-thinking-mindset -->

Critical Thinking Mindset — Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act.

Anti-hallucination: Never present guess as fact — cite sources for every claim, admit uncertainty freely, self-check output for errors, cross-reference independently, stay skeptical of own confidence — certainty without evidence root of all hallucination.

<!-- /SYNC:critical-thinking-mindset -->

Closing Reminders

IMPORTANT MUST ATTENTION Goal: Plan and generate a developer KPI-style quality-work report from local git history only.

Protocols in force (concise digest of the SYNC/shared blocks this skill carries):

• Critical Thinking: trace every KPI/value claim; confidence >80% to act.
• AI Mistake Prevention: verify generated content against evidence, trace downstream references, verify all affected outputs, re-read after context loss, surface ambiguity.

<!-- SYNC:critical-thinking-mindset:reminder -->

MUST ATTENTION apply critical + sequential thinking — every claim needs appropriate traced evidence (file:line for repo/code claims; source URL or artifact section for research, product, content, and docs claims); confidence >80% to act, <60% DO NOT recommend. Anti-hallucination: never present guess as fact, admit uncertainty freely, cross-reference independently, stay skeptical of own confidence.

<!-- /SYNC:critical-thinking-mindset:reminder -->

IMPORTANT MUST ATTENTION use local git history only. IMPORTANT MUST ATTENTION trigger planning before qualitative analysis; this is a large task. IMPORTANT MUST ATTENTION default to develop, fallback to main, and use last 60 days when the user does not specify. IMPORTANT MUST ATTENTION do not present authored or integration signals as complete measures of human performance. IMPORTANT MUST ATTENTION shared feature-branch implementation credit follows direct commit authors, not merge authors; never let raw churn or zero-change merge/admin commits inflate implementation SP or man-day estimates. IMPORTANT MUST ATTENTION never publish a single ambiguous MD number; show no-AI and AI-assisted MD separately.

<!-- SYNC:ai-mistake-prevention:reminder -->

MUST ATTENTION apply AI mistake prevention — verify generated content against evidence, trace downstream references before deleting or renaming, verify all affected outputs, re-read files after context loss, and surface ambiguity before acting.

<!-- /SYNC:ai-mistake-prevention:reminder -->

<!-- CODEX:SYNC-PROMPT-PROTOCOLS:START -->

Hookless Prompt Protocol Mirror (Auto-Synced)

Source: .claude/.ck.json + .claude/skills/shared/sync-inline-versions.md (:full blocks) + .claude/scripts/lib/hookless-prompt-protocol.cjs

[WORKFLOW-EXECUTION-PROTOCOL] [BLOCKING] Workflow Execution Protocol — MANDATORY IMPORTANT MUST CRITICAL. Do not skip for any reason.

Generic portability boundary: Reusable skills and protocol text stay project-neutral; project-specific conventions are discovered from docs/project-config.json and docs/project-reference/. Apply shared AI-SDD from shared/sdd-artifact-contract.md. Read docs/project-config.json and docs/project-reference/docs-index-reference.md, then open the project reference docs named there. For spec, test-case, behavior-change, public-contract, or docs/specs/ work, route through the local spec docs named by the docs index: feature-spec-reference.md, spec-system-reference.md, spec-principles.md, and workflow-spec-test-code-cycle-reference.md when specs/tests/code must stay synchronized. If either file or a required reference doc is missing or stale, auto-run $project-init (or the narrow lower-level route such as $project-config, $docs-init, $scan-all, or $scan --target=<key>) before ordinary project-specific work. Any supported AI tool may execute when this shared context and local docs are available.

1. DETECT: If the prompt starts with an explicit slash skill/workflow command, execute it directly. Otherwise match the prompt against the workflow catalog and skill list. 2. ANALYZE: Choose the best option: execute directly, invoke a skill, activate a standard workflow, or compose a custom step combination. 3. AUTO-SELECT: Pick the best option yourself. Do not ask the user to choose between direct execution, skill, standard workflow, or custom workflow. 4. ACTIVATE: For a selected workflow, call $start-workflow <workflowId>; for a selected skill, invoke that skill; for a custom workflow, sequence custom steps directly; for direct execution, proceed with the task. 5. CREATE TASKS: task tracking for ALL workflow/skill/custom steps before execution when the selected path has multiple steps. 6. EXECUTE: Advance per the Workflow Step Advancement & Parallel Phases rule in your context instructions — model-driven; a sub-agent completion advances a step identically to an inline call; a parallel-phase group is an all-return barrier (advance only after ALL members return, never serialize it)

Shared AI-SDD Protocol Markers

Source: .claude/skills/shared/sync-inline-versions.md

SYNC:ai-sdd-artifact-contract

AI-SDD Artifact Contract — Shared spec-driven development rules stay portable and source-owned.

>

1. Keep reusable AI-SDD principles in .claude; put repository-specific paths, commands, owners, products, and formats in project config/reference docs.

2. Preserve cycle: spec -> plan -> tasks -> implement -> verify -> update spec/docs.

3. Trace every requirement or invariant through decision, task, TC/test, source evidence, and docs/spec update.

4. Treat code-to-spec extraction as reference-only until accepted by the canonical spec owner.

5. Any supported AI tool may plan, implement, review, or verify with synced context; using multiple tools is optional.

6. Update .claude source first, then sync generated mirrors; do not manually edit .agents, .codex, or AGENTS.md. — why: mirrors are generated artifacts; hand-edits are overwritten on the next sync

7. If docs/project-config.json, root instruction files, or a required project-reference doc is missing or stale, auto-run $project-init or the narrow lower-level route before ordinary project-specific work.

>

Active reference: shared/sdd-artifact-contract.md in the active skills root.

---

SYNC:ai-sdd-artifact-contract:reminder

• MANDATORY Apply shared/sdd-artifact-contract.md; keep reusable AI-SDD in .claude and local rules in project docs.
• MANDATORY Code-to-spec extraction is reference-only until canonical acceptance; any supported AI tool may execute with synced context.
• MANDATORY Update .claude source before syncing generated mirrors; do not manually edit .agents, .codex, or AGENTS.md.
• MANDATORY Missing or stale project config, root instruction files, or required reference docs route project-specific work through $project-init or the narrow setup route automatically.

[TASK-PLANNING] [MANDATORY] BEFORE executing any workflow or skill step, create/update task tracking for all planned steps, then keep it synchronized as each step starts/completes.

[LESSON-LEARNED-REMINDER] [BLOCKING] Task Planning & Continuous Improvement — MANDATORY. Do not skip.

Break work into small tasks (task tracking) before starting. Add final task: "Analyze AI mistakes & lessons learned".

Extract lessons — ROOT CAUSE ONLY, not symptom fixes:

1. Name the FAILURE MODE (reasoning/assumption failure), not symptom — "assumed API existed without reading source" not "used wrong enum value". 2. Generality test: does this failure mode apply to ≥3 contexts/codebases? If not, abstract one level up. 3. Write as a universal rule — strip project-specific names/paths/classes. Useful on any codebase. 4. Consolidate: multiple mistakes sharing one failure mode → ONE lesson. 5. Recurrence gate: "Would this recur in future session WITHOUT this reminder?" — No → skip $learn. 6. Auto-fix gate: "Could $code-review/$code-simplifier/$security-review/$lint catch this?" — Yes → improve review skill instead. 7. BOTH gates pass → ask user to run $learn. [CRITICAL-THINKING-MINDSET] Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act. Anti-hallucination principle: Never present guess as fact — cite sources for every claim, admit uncertainty freely, self-check output for errors, cross-reference independently, stay skeptical of own confidence — certainty without evidence root of all hallucination. AI Attention principle (Primacy-Recency): Put the 3 most critical rules at both top and bottom of long prompts/protocols so instruction adherence survives long context windows. Goal-driven execution: Define success criteria first, loop until verified, and stop only when observable checks pass. Tests verify intent: Tests must protect business rules/invariants and fail when the protected intent breaks, not only mirror current behavior.

Common AI Mistake Prevention (System Lessons)

• Re-read files after context compaction. Edit requires prior Read in same context; compaction wipes read state. Re-read before editing.
• Grep for old terms after bulk replacements. AI over-trusts find/replace completeness. Grep full repo after bulk edits for missed refs in docs/configs/catalogs.
• Check downstream references before deleting. Deletions cascade doc/code staleness. Map referencing files before removal.
• After memory loss, check existing state before creating new. Compaction wipes prior-work memory. Query current state to resume — never blindly duplicate.
• Verify AI-generated content against actual code. AI hallucinates APIs, class names, method signatures. Grep to confirm existence before documenting/referencing.
• Trace full dependency chain after edits. Changing a definition misses downstream consumers. Trace the full chain.
• When renaming, grep ALL consumer file types. Some file types silently ignore missing refs (no compile error). Search code, templates, configs, generated files.
• Trace ALL code paths when verifying correctness. Code existing ≠ code executing. Trace early exits, error branches, conditional skips — not just happy path.
• Update docs that embed canonical data when source changes. Docs inlining derived data (workflows, schemas, configs) go stale silently. Update all embedding docs alongside source.
• Verify sub-agent results after context recovery. Background agents may finish while parent compacted — grep-verify output, don't trust assumed completion.
• Cross-check full target list against sub-agent assignments. Parallel sub-agents by category miss boundary items. Reconcile union of assignments against target list before proceeding.
• Sub-agents inherit knowledge only from their agent .md definition — use custom agent types, not built-in Explore. Tool adoption = permission + knowledge + enforcement (numbered workflow step).
• Persist sub-agent findings incrementally, not as a final batch. Long sub-agents hit cutoffs before final write — findings lost. Instruct append-per-section to report file.
• When debugging, ask "whose responsibility?" before fixing. Trace caller (wrong data) vs callee (wrong handling). Fix at responsible layer — never patch symptom site.
• Grep ALL removed names after extraction/refactoring. Primary file "done" ≠ secondary files clean. Grep entire scope for every removed symbol before declaring complete.
• Assume existing values are intentional — ask WHY before changing. Pattern-matching as "wrong" skips context. Before changing any constant/limit/flag: read comments, git blame, surrounding code.
• Verify ALL affected outputs, not just the first. One build green ≠ all green. Multi-stack changes (backend/frontend/tests/docs) require verifying EVERY output.
• Evaluate fit before copying a nearby pattern. Closest example ≠ matching preconditions — verify the new context shares the same constraints, base classes, scope, lifetime.
• Holistic-first debugging — resist nearest-attention trap. Don't dive into first plausible cause. List EVERY precondition (config, env vars, paths, DB, endpoints, creds, versions, DI, data). Verify each against evidence (grep/query — not reasoning). Ask "what would falsify this?" — if nothing, it's not a hypothesis. Most expensive failure: going deeper in "obvious" layer while bug sits in layer never questioned.
• Surgical changes — apply the diff test (context-aware). Two modes: (1) Bug fix → every line traces to the bug; no restyling; orphan cleanup only for imports YOUR changes made unused. (2) Review/enhancement → implement improvements AND announce as "Enhancement beyond main request: [what]". Never silently scope-creep. Diff test: "Would this line exist if I wasn't asked to do X?" — if no, delete or announce.
• Surface ambiguity before coding — don't pick silently. Multiple valid interpretations → present each with effort: "[Request] could mean (1) [N h], (2) [N h]. Which matters?" List scope/format/volume/constraints assumptions first. If simpler path exists, say so. Never silently pick.
• [MANDATORY FIRST ACTION] ALWAYS activate a suitable skill or workflow BEFORE responding. Match task against workflow catalog + skill list; invoke via skill invocation or $start-workflow <workflowId>. NEVER answer or write code before checking. Skip = protocol violation.
• Why-Review adversarial mindset — apply when reviewing any plan, decision, or design. Default SKEPTIC not VALIDATOR: steel-man a rejected alternative, invert each stated reason ("what does it sacrifice?"), stress-test top 2-3 assumptions, run pre-mortem ("ships, fails in 3 months — what breaks?"), surface 1-2 alternatives author missed. Section presence ≠ quality; quality = causal reasoning + concrete mitigations + evidence, not "it's better" or "monitor closely".
• Front-load report-write in sub-agent prompts for large reviews. Many-file sub-agents hit budget before final write — findings lost. Design prompts so: (1) report-write is first explicit deliverable, (2) append per-file/section (not batched), (3) scope bounded so reads don't exhaust budget. Truncated mid-sentence with no report file → spawn narrower scope, don't retry same prompt.
• After context compaction, re-verify all prior phase outcomes before continuing. Summaries describe intent, not environment state (git index, filesystem, processes). On resume, FIRST audit: git status, re-read modified files, verify filesystem. Every "completed" claim is an untested hypothesis until evidence confirms.
• OOM/memory: check row count before row size. Triage: (1) Unbounded query — no DB filter for trigger? Push filter to DB; eliminates OOM. (2) Large rows? Projection reduces proportionally. Row reduction > projection in ROI.
• Keep domain concepts out of generic/shared/infrastructure layers. Reusable layer (shared library, framework, infra module) must reference NO consumer-specific domain concept — tenant/customer/product IDs, business entities, feature rules. Leak compiles + runs → passes review silently while coupling the "reusable" layer to one consumer. Keep shared type domain-free; push domain fields/logic down into the consumer via subclass/composition. — why: a layer coupled to one consumer's domain is no longer reusable.

<!-- CODEX:SYNC-PROMPT-PROTOCOLS:END -->

Developer Quality Work Analysis Workflow

Use after running scripts/git-developer-performance.cjs.

Required Sequence

1. Read analysis-plan.md. 2. Count distinct contributors from the plan/data and create one todo task per contributor. Contributors are person-consolidated by identity map, email, and high-confidence aliases unless an identity map overrides them. 3. For many contributors, split disjoint developer lists across subagents; merge findings afterward. 4. For each contributor, read work-packets/{developer}.md. 5. Inspect representative commits with git show --stat --find-renames <hash>. 6. Inspect patches for high-impact, unclear, risky, or quality-related commits. 7. Distinguish direct authored work from merge/admin integration work. 8. For shared feature branches, attribute implementation value to each developer's own direct commits, not to the merge author or PR owner. 9. Treat zero-file merge/admin commits as integration signal only; do not add them to implementation SP. 10. Discount generated, migration-designer, docs/spec, i18n-sorting, lockfile, and repeated follow-up churn before estimating. 11. For rechecks or velocity mismatch, write an external analysis file first and synthesize each contributor's direct work as one "giant commit." 12. Split the giant commit into atomic 1/2/3/5/8/13 SP clusters; displayed theme totals over 13 SP must be explicit sums, not one unsplit story. 13. Separate product/domain delivery, infrastructure/tooling, docs/generated churn, and merge/admin integration before producing team velocity totals. 14. Synthesize value; do not stop at commit counts, file counts, or churn. 15. Estimate both no-AI (man_days_traditional) and AI-assisted (man_days_ai) ranges. 16. Run a velocity sanity check against active days and selected period; re-audit implausible SP/man-day ranges. 17. Write per-contributor synthesis to analysis/{developer}.md. 18. Write quality-work-summary.md and evidence-proof.md.

Work Cluster Estimation

Cluster commits by intent, module, date proximity, and subject.

• 1 SP: small isolated change, docs tweak, simple config, narrow fix.
• 2 SP: small feature/fix touching a few files.
• 3 SP: moderate change across files/modules with clear business value.
• 5 SP: complex feature, refactor, bug investigation, or integration impact.
• 8 SP: large multi-module work with non-trivial test/integration risk.
• 13 SP: very large or ambiguous work; split if possible.

Man-days: report both ranges. man_days_traditional is no-AI baseline: 3-5yr developer, 6 productive hours/day. man_days_ai assumes an AI coding assistant plus project context; apply plan-skill AI speedup (SP 1 about 2x, 2-3 about 3x, 5-8 about 4x, 13+ about 5x) with review overhead.

Anti-inflation guardrails: estimate authored implementation work first. In no-ff multi-author feature branch merges, implementation follows the original direct commit authors. Merge/admin work is separate unless the merge commit itself has substantive resolved changes. Generated files, EF designer snapshots, docs/specs, i18n sorting, lockfiles, and repeated feedback/fix commits must reduce confidence or be discounted.

Velocity sanity: when estimates feel too high or low, do not tune numbers by intuition. Re-open representative patches, synthesize the contributor as one giant commit, split into atomic clusters, and then revise.

KPI-Style Evaluation

Use evidence-backed dimensions, not opaque scores:

• Delivery value: user-facing features, workflow enablement, integration impact.
• Bug-fix value: defect severity, root-cause depth, regression risk reduced.
• Refactor value: simplification, maintainability, migration, performance, infrastructure quality.
• Quality value: tests, observability, docs, error handling, safer APIs, deleted dead code.
• Risk/caveat: generated/bulk changes, revert churn, unclear ownership, missing local history.

Per-Contributor Output

Include:

• Distinct identity and commit counts.
• Consolidated aliases in originalIdentities, when present.
• Direct commits vs merge/admin commits.
• Work clusters with story points, man_days_traditional, and man_days_ai.
• Proof basis for each estimate: direct commit hashes, changed paths, discounted churn, and confidence.
• Value contributed: feature/change, bug fix, refactor, quality/test/docs, integration/admin.
• Code-quality impact: improved maintainability, risk, tests, generated/bulk changes, risky churn.
• Evidence: commit hashes and changed paths.
• Confidence and caveats.

Final Summary Output

Include:

• Total distinct contributors.
• Total estimated story points, no-AI man-days, and AI-assisted man-days.
• Top contributed values, not only commit counts.
• Separate authored work from merge/admin integration.
• Avoid ranking by commits/lines alone; compare only evidence-backed value.
• Include a velocity sanity note when estimates are high relative to active days or team expectations.
• Warnings for identity splits, bulk changes, squashed commits, generated changes, and low-confidence estimates.