Agent skill

Incremental Project Analysis

by devcodex-labs in devcodex-labs/devcodex

增量项目分析 Owner — 知识快照、内容 digest、分析视角覆盖、changed→impact 选择性失效、智能分批可见交付、全局优先级综合、分析精度合同与双层验证;大型/逐文件分析禁止只缓存自然语言总结。

AGPL-3.0Auto-check passed

Install Incremental Project Analysis

skills CLI
$ npx skills add devcodex-labs/devcodex --skill incremental-project-analysis -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install devcodex-labs/devcodex incremental-project-analysis --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/devcodex-labs/devcodex.git skills-src && mkdir -p .claude/skills && cp -r skills-src/content/skills/incremental-project-analysis .claude/skills/incremental-project-analysis && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
incremental-project-analysis
GitHub stars
439
Token cost
~2k tokens
SKILL.md length
597 words
Files
3
Skills in repo
70
Repo updated
First seen
Licence
AGPL-3.0

At a glance

增量项目分析 Owner — 知识快照、内容 digest、分析视角覆盖、changed→impact 选择性失效、智能分批可见交付、全局优先级综合、分析精度合同与双层验证;大型/逐文件分析禁止只缓存自然语言总结。

  • Works in 6 steps: 首次分析冻结:repo/base tree(或等价)、排除边界、文件… → 首次无快照时执行 deterministic observe → batch… → 稳定事实层(symbols、imports、配置锚点)按 content… → …
  • SKILL.md covers 职责, 何时触发, 与相邻 Skill 分界 and 核心产物, plus 4 more sections
  • Calls git

What it does

Incremental Project Analysis is an agent skill from devcodex-labs/devcodex. 增量项目分析 Owner — 知识快照、内容 digest、分析视角覆盖、changed→impact 选择性失效、智能分批可见交付、全局优先级综合、分析精度合同与双层验证;大型/逐文件分析禁止只缓存自然语言总结。

Its SKILL.md is about 2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 3 other files (for example `agents/openai.yaml` and `intent.json`).

The repository describes itself as: Intent-driven AI coding workflow runtime for consistent context, skills, approvals, validation, and handoffs across six AI coding hosts. The licence is AGPL-3.0.

Example prompts

  • “/incremental-project-analysis”

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

  1. 首次分析冻结:repo/base tree(或等价)、排除边界、文件 contentDigest、coverageLevel、依赖边、policyVersion。
  2. 首次无快照时执行 deterministic observe → batch validate → bootstrap;只保存从当前字节可重建的结构声明,不得把 bootstrap 宣称为人工深读。
  3. 稳定事实层(symbols、imports、配置锚点)按 content identity + rangeDigest 复用;职责、风险、建议只能以 agent-semantic authority 单独接受。
  4. 判断层绑定 analysisLens / dependsOn / policyVersion;代码未变但目标变时只补 lens-gap。
  5. ProjectKnowledgeBindingV1 的 repo/root、配置、解析器、测试路线或 Profile 身份错配时强制 full-required;inventory 内容差异走 delta/impact,不因正常小变化误判为环境错配。
  6. 禁止仅保存 Markdown 自然语言总结当作可复用知识。

What it can do on your machine

Read from SKILL.md and the folder at commit 1dd4525. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • git

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Incremental Project Analysis loads about 2k tokens when it runs. Until then it costs about 34 tokens; SKILL.md has 597 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~34
When it runs · the whole SKILL.md, loaded when a task matches
~2k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from devcodex-labs/devcodex at commit 1dd4525, republished under its AGPL-3.0 licence (© devcodex-labs). 597 words, ~1,993 tokens.

Download SKILL.mdSave it as .claude/skills/incremental-project-analysis/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.
name
incremental-project-analysis
description
增量项目分析 Owner — 知识快照、内容 digest、分析视角覆盖、changed→impact 选择性失效、智能分批可见交付、全局优先级综合、分析精度合同与双层验证;大型/逐文件分析禁止只缓存自然语言总结。

Incremental Project Analysis

职责

承接大库 / 逐文件 / 多会话项目分析的可复用认知层:把稳定事实、依赖图、分析视角覆盖与结论新鲜度做成可校验快照,支持增量重算、分批交付与全局综合。
不替代 skill-gap-analysis 的规模分批扫描本身;不替代 analyze-default 的只读多轮分析主流程。本 Skill 提供快照、精度、批次交付与验证状态机。

何时触发

  • 用户要求完整深度 / 逐文件 / 多批项目分析,且语料 non-small。
  • 用户要求「后续只分析增量 / 不要每轮重读全库」。
  • analyze-default / skill-gap-analysis 在 ScaleDecision 为 batched / sampled+deep-read 时必须调用本 Skill 的交付与精度门禁。
  • 用户要求最终确认清单 / 可吸纳包时,叠加 FindingThemeCoverageMatrix(与 report ABS-17 一致)。

与相邻 Skill 分界

Skill本 Skill 管对方管
skill-gap-analysis知识快照、全局 backlog、精度与验证阶段规模路由、缺口 vs Owner 矩阵、corpus inventory
analyze-default快照复用、BatchProgress、GlobalBacklog、双产物三轮收敛、PCV、只读边界、治理 intake
memory快照路径链接;禁止把快照正文写入 SUMMARY会话日记、handoff、TimingCard
reportTheme+Detail 链接、CoverageMatrix 字段报告路径与五项验证列

核心产物

产物说明
ProjectKnowledgeSnapshotV2可持久化 accepted 项目知识;含 repo/root、Merkle、环境绑定、policy/schema 版本
FileKnowledgeRecordV2每文件 content identity、coverageLevel、结构观察与已接受声明
SemanticClaimV1类型化声明;绑定 sourceRange/rangeDigest、authority、lens、policy 与依赖
ProjectKnowledgeBindingV1绑定唯一 repo/root、inventory Merkle、配置、解析器、测试路线与 Profile 身份
ImpactGraphV1依赖 / 消费者 / 配置 / 契约边
AnalysisLensRecordV1分析视角覆盖(lens / questionFingerprint)
IncrementalAnalysisPlanV2本轮只读集合:binding + changed + impact + lens-gap + deterministic sample
IncrementalAnalysisReceiptV2复用/重算/失效分类、抽样复证与 accepted pointer 结果
BatchProgressCard每 accepted 批用户可见交付
GlobalOptimizationBacklogV1全批后唯一高/中/低优先级清单
ValidationPlanV1 / BatchValidationResultV1 / GlobalValidationResultV1双层验证

落盘建议:<active-root>/reports/analysis/<agent>/YYYYMMDD/deep/ 或任务目录 artifacts/knowledge-snapshot/;禁止写入 Agent SUMMARY 正文。

源仓 runtime Owner 为 scripts/lib/project-knowledge-store.js,内部 CLI 为 scripts/project-analysis-state.js status|plan|observe|bootstrap|accept。accepted runtime 固定落到 <active-root>/.runtime-state/project-knowledge/v2/<repoId>/snapshot.json;status、plan、observe 必须零写入,bootstrap 只生成可复证的 content-structured 首轮基线,只有 BatchValidationResultV1=pass 且 sample oracle=pass 的 bootstrap/accept 才可在所有批次内存验收后推进一次 runtime pointer 并写任务证据。

V1 只读兼容:v1/<repoId>/snapshot.json 仅用于返回 compatibility-v1 + migrationRequired 导航信息,禁止作为 V2 reuse/accept 基线;下一次 bootstrap 必须走 full-required 并生成独立 V2 accepted snapshot。禁止迁移时原地改写 V1。

当 execution optimization 为 full-only,或 project-knowledge-reuse 的 ExecutionOptimizationFeatureDecisionV1 为 off / shadow / rolled-back / sunset,或 state/snapshot schema/identity 无效、oracle 失败时,plan 必须返回 full-project-analysis,不得复用 snapshot records;status 同时公开 executionOptimizationMode、feature decision 与 reuseAllowed=false。kill switch 只关闭加速,不跳过 inventory、智能分批、逐批验证、最终全局验证或高/中/低 backlog。


Gate 索引(执行正文)

ProjectKnowledgeSnapshotGate(ABS-01)
  1. 首次分析冻结:repo/base tree(或等价)、排除边界、文件 contentDigest、coverageLevel、依赖边、policyVersion。
  2. 首次无快照时执行 deterministic observe → batch validate → bootstrap;只保存从当前字节可重建的结构声明,不得把 bootstrap 宣称为人工深读。
  3. 稳定事实层(symbols、imports、配置锚点)按 content identity + rangeDigest 复用;职责、风险、建议只能以 agent-semantic authority 单独接受。
  4. 判断层绑定 analysisLens / dependsOn / policyVersion;代码未变但目标变时只补 lens-gap。
  5. ProjectKnowledgeBindingV1 的 repo/root、配置、解析器、测试路线或 Profile 身份错配时强制 full-required;inventory 内容差异走 delta/impact,不因正常小变化误判为环境错配。
  6. 禁止仅保存 Markdown 自然语言总结当作可复用知识。

生产 CLI 必须从当前 bounded inventory 的真实字节构建 ImpactGraphV1:首期至少解析 JS/TS 静态相对 require/import/export/dynamic import literal 与 Markdown 本地链接,记录 typed edge、source、evidenceStrength、覆盖率、未解析引用和 unknownConsumerPaths。图必须带 builderVersion;旧/未知 builder 首次迁移 full-required,之后普通图拓扑变化使用当前图重新计算影响闭包,不得仅因 graph identity 改变就无条件全文重读。

SemanticClaimBoundaryGate
  1. 每条可复用声明必须是 SemanticClaimV1,至少包含 type / statement / authority / sourceRange / rangeIdentity / sourceContentDigest / lensId / policyVersion / dependsOn / claimDigest / status。
  2. inventory-only 只能声明文件存在与内容身份;content-structured 只能声明文本中直接可解析的 heading、symbol、import、config/test 结构;只有 agent-semantic 可声明职责、风险或建议。
  3. rangeDigest 必须从当前文件精确行范围重算;path、content、range、lens 或 claim digest 任一不一致即拒绝该批,禁止“修补后继续复用”。
  4. accepted snapshot 只持久化 status=accepted 声明;candidate/provisional/invalid 只能留在当轮候选输出,不得推进 runtime pointer。
SelectiveInvalidationGate
  1. 变化集优先 git diff --name-status,并用当前 digest 覆盖未提交/未跟踪/rename。
  2. 直接变化扩展到依赖、消费者、公共契约、配置、生成链、测试与文档闭包。
  3. 状态:fresh / stale / coverage-gap / incompatible / full-required。
  4. 强制升级:快照损坏、base 不可达、schema/builder 不兼容、高风险变更、闭包过大、抽样复证失败;动态依赖消费者自身发生变化时 full-required,其他文件变化时必须把全部 unknown consumer 注入 affected closure 并显式重读,禁止静默复用。

mtime/size 只能做候选加速;fresh 必须由 current bytes content identity 证明。rename 必须同时写旧 path tombstone 与新 path identity,delete 写 tombstone;配置变化保守扩到受影响全域。ImpactGraph 每条边必须含 type/source/evidenceStrength,图覆盖不足时不得声称 selective complete;unknownConsumerPaths 与 builder/stats 必须进入 graph identity。graph identity 变化会使未完成旧 plan 失效,但只在 builder 迁移或其他升级条件命中时触发全文;正常拓扑变化应以新图计算 changed/impact closure。

Show full SKILL.md (231 more words)Show less
AdaptiveBatchDeliveryGate(ABS-02)
  1. 按模块 / namespace / 消费者 / 风险分批;每批有 budget 与 checkpoint。
  2. 每个 accepted 批次必须立即输出并持久化:BatchProgressCard + 本批发现 + checkpoint + snapshot delta。
  3. 禁止等待全量完成才首次交付用户可见结果。
  4. 恢复只从最后一个 accepted 批次继续;invalid/blocked 不得推进指针。

复用抽样必须按 contentId/path 稳定排序选择 5%,最少 3、最多 20,不足 3 全抽。任一 content/fact mismatch 使本批 invalid + full-required;历史 runtime 只标 stale,不自动删除。

observe 与 bootstrap 的候选选择、结构提取、claimId、rangeDigest、Merkle root 和 sample 必须确定性;同一 repo/root、字节、环境身份与 lens 重跑必须得到相同 identity。持久化顺序为“先验证全部身份与任务证据,最后单次原子替换 runtime pointer”;任务证据写入失败时 pointer 不得推进。

GlobalPrioritySynthesisGate(ABS-03)
  1. 全部计划批次 accepted 且 CRS/PCV 收敛后,跨批去重合并,输出唯一 GlobalOptimizationBacklogV1(高/中/低 + 推荐顺序)。
  2. 未完成 / invalid / blocked → 只能 partial/provisional,禁止 final/completed 宣称。
  3. findingId 与 priority 分字段(ABS-05),禁止用 P1 编号冒充 P0 优先级。
AnalysisPrecisionContractGate(ABS-04)
  1. 开批前冻结 depthTier:deep | standard | light。
  2. 每文件笔记必有 evidenceStrength:agent-semantic | content-structured | inventory-only。
  3. 交付必须 ThemePriorityBacklog + FindingDetailLedger(去噪 high/medium;默认排除纯归档 low)。
  4. claimBoundary:inventory/content-structured 不得宣称「人工长文精读」;主题 N 条 不得宣称「仅有 N 个问题」。
DualLayerValidationGate

生命周期:

text
plan → execute → batch validate → accept/deliver → synthesize → global validate → final
规则说明
执行前冻结 ValidationPlanV1(风险、声明、消费者、证据、升级路线)
批次仅 BatchValidationResult=pass 可 accepted / 写 checkpoint / snapshot delta
全局仅 GlobalValidationResult=pass 可 final backlog / completed
失败fail/inconclusive 隔离候选输出;按 targeted→related→full 扩大或重做;全局失败保持 provisional

执行清单(最小)

  1. 确认 ScaleDecision(委托 skill-gap-analysis)。
  2. 加载 V2 snapshot;若仅有 V1,标记只读兼容并规划 full-required bootstrap。
  3. 冻结 depthTier + lens + ValidationPlan。
  4. 构建 ProjectKnowledgeBindingV1,计算 changed / impact / lens-gap → IncrementalAnalysisPlanV2。
  5. 分批执行:observe/语义补充 → claim range 复证 → batch validate → accepted ProgressCard;首次基线用 bootstrap,后续按差量 accept。
  6. 全批后 CRS/PCV → GlobalBacklog → global validate。
  7. 报告双产物 + CoverageMatrix(确认清单场景)+ TimingCard(长任务)。

反模式

  • 每轮全文重读却不更新 digest / 失效图。
  • 只交付主题 8 条,隐瞒 high/medium 明细。
  • 把 waiting-user 时间算进「分析执行慢」且无 TimingCard。
  • 在 SUMMARY 粘贴快照正文。
  • 未 batch validate 就写 accepted checkpoint。
  • 把 V1 snapshot 静默升级、继续复用或写回原路径。
  • 把结构提取结果描述成“逐文件人工深读完成”或“问题只有 N 个”。

验证路线

正负向至少覆盖:首轮 observe/bootstrap 确定性;单文件变更只重算闭包;配置变更使未改源码失效;新 lens 触发 coverage-gap;rename/tombstone;repo/root/config/parser/test/Profile/content/range identity 错配;V1 只读兼容;5% sample 确定性;candidate/provisional 不持久化;accepted 缺 ProgressCard 失败;global 未完成不得 final;findingId/priority 混用失败;claim authority 越界失败。

© devcodex-labs, AGPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 2 other files in content/skills/incremental-project-analysis of devcodex-labs/devcodex.

  • SKILL.md
  • agents/openai.yaml
  • intent.json

Open the folder on GitHubat commit 1dd4525

Compare with similar skills

Incremental Project Analysis next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Incremental Project Analysis compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Incremental Project Analysis this skilldevcodex-labs/devcodex439—~2kAutomated safety check: PassAGPL-3.0
Implement Changepnpm/pnpm37k—~1.1kAutomated safety check: PassMIT
Incremental Implementationaddyosmani/agent-skills102k1 repos~2.3kAutomated safety check: PassMIT
Make Changesremix-run/remix33k—~2.4kAutomated safety check: PassMIT
Orch Change Featureaffaan-m/ECC274k1 repos~420Automated safety check: PassMIT
Morning Digestyc-software/qm15k—~754Automated safety check: PassMIT

Similar skills

  • Implement a pnpm feature, bug fix, or refactor by checking existing capabilities, prioritizing code reuse and deduplication, assessing architecture impact, and validating the final change.

    37k GitHub stars~1.1k tokensUpdated today
    DevelopmentAuto-check passed
  • Incremental Implementation

    addyosmani/agent-skills

    Delivers a change in thin vertical slices, each implemented, tested, verified and committed before the next, using vertical, contract-first or risk-first slicing.

    102k GitHub starsUsed in 1 repo~2.3k tokens
    Agent WorkflowsAuto-check passed
  • Make Changes

    remix-run/remix

    Create or update Remix repo change files under packages//.changes.

    33k GitHub stars~2.4k tokensUpdated yesterday
    DevelopmentAuto-check passed
  • Orchestrate altering an existing, working feature to new desired behavior — update its tests to the new spec, change the implementation to match, review, and gated commit.

    274k GitHub starsUsed in 1 repo~420 tokens
    Auto-check passed
  • Morning Digest

    yc-software/qm

    Assemble and deliver one short morning digest for the person who owns the cron — what changed overnight in the sources they already connected, and what it means for their day.

    15k GitHub stars~754 tokensUpdated today
    Productivity & AutomationAuto-check passed
  • Change Management

    sickn33/agentic-awesome-skills

    Implement change management processes. An agent skill from sickn33/agentic-awesome-skills.

    47k GitHub starsUsed in 2 repos~3.5k tokens
    DevOps & CloudAuto-check passed

More from devcodex-labs/devcodex

All 70 skills in this repo
  • Accessibility I18n

    devcodex-labs/devcodex

    无障碍与国际化专家 Owner — 当任务涉及可访问性、键盘操作、焦点、屏幕阅读器、ARIA、语言地区、本地化、RTL、翻译资源、用户可见文案或多语言文档时使用;要求把包容性体验和本地化验证绑定到真实用户路径。

    439 GitHub stars~718 tokensUpdated 20 days ago
    Auto-check passed
  • AI Agent System Architecture

    devcodex-labs/devcodex

    AI Agent 系统架构专家 Owner — 当任务涉及 Agent 路由、工具调用、上下文管理、记忆、状态机、权限、人机协作、可观测性、回放验证或模型辅助治理时使用;要求把 Agent 行为设计成可解释、可恢复、可审计。

    439 GitHub stars~2.4k tokensUpdated 20 days ago
    Auto-check passed
  • API Contract Architecture

    devcodex-labs/devcodex

    API 契约架构专家 Owner — 当任务涉及 public API、HTTP/SDK/CLI 契约、版本兼容、错误模型、分页过滤、幂等、Schema、类型、迁移或消费者影响时使用;要求先冻结消费者契约,再设计实现与验证。

    439 GitHub stars~865 tokensUpdated 20 days ago
    Auto-check passed
  • Architecture Design

    devcodex-labs/devcodex

    架构设计文档编排 Owner — 当用户要求架构设计、系统设计、技术架构或可指导开发、Review 与任务拆分的完整方案时使用;要求从业务流程反推节点、状态、数据、一致性、异常补偿、ADR 与实施任务。

    439 GitHub stars~1.1k tokensUpdated 20 days ago
    Auto-check passed
  • Audit Common

    devcodex-labs/devcodex

    审查公共维度 G0~G5 + Profile Freshness Check — 所有 audit 子类型必先执行的基础维度层

    439 GitHub stars~4.1k tokensUpdated 20 days ago
    Auto-check passed
  • Audit Session

    devcodex-labs/devcodex

    审计工作流的跨会话状态机 — 在 <audit-root/.audit-state/<session-id.json 持久化轮次/发现项/收敛状态,支持 Token 中断后精准恢复

    439 GitHub stars~1.8k tokensUpdated 20 days ago
    Auto-check passed

Questions about Incremental Project Analysis

What does Incremental Project Analysis do?

增量项目分析 Owner — 知识快照、内容 digest、分析视角覆盖、changed→impact 选择性失效、智能分批可见交付、全局优先级综合、分析精度合同与双层验证;大型/逐文件分析禁止只缓存自然语言总结。. Incremental Project Analysis is an agent skill from devcodex-labs/devcodex.

How do I install Incremental Project Analysis in Claude Code?

Run `npx skills add devcodex-labs/devcodex --skill incremental-project-analysis -a claude-code`. Or copy the skill folder (content/skills/incremental-project-analysis in devcodex-labs/devcodex) into .claude/skills/incremental-project-analysis in your project. Claude Code loads it when a task matches its description.

How do I install Incremental Project Analysis in Codex?

Run `npx skills add devcodex-labs/devcodex --skill incremental-project-analysis -a codex`. Or copy the skill folder (content/skills/incremental-project-analysis in devcodex-labs/devcodex) into .agents/skills/incremental-project-analysis in your project. Codex loads it when a task matches its description.

Can I use Incremental Project Analysis in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add devcodex-labs/devcodex --skill incremental-project-analysis -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/incremental-project-analysis, .gemini/skills/incremental-project-analysis, .github/skills/incremental-project-analysis and .opencode/skills/incremental-project-analysis in your project.

What does Incremental Project Analysis need to run?

Going by SKILL.md and its folder, Incremental Project Analysis needs the command-line tools its instructions call (git).

Does Incremental Project Analysis access the network?

SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Incremental Project Analysis safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Incremental Project Analysis use?

Incremental Project Analysis is published under the AGPL-3.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Incremental Project Analysis use?

About 2k tokens (SKILL.md is roughly 8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Incremental Project Analysis?

Skills that share tags, products or a category with Incremental Project Analysis: Implement Change (pnpm/pnpm, 37k stars), Incremental Implementation (addyosmani/agent-skills, 102k stars), Make Changes (remix-run/remix, 33k stars) and Orch Change Feature (affaan-m/ECC, 274k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Incremental Project Analysis?

devcodex-labs (a GitHub organization) maintains it in devcodex-labs/devcodex, which has 439 GitHub stars. The repository holds 70 skills in this directory. The repository was last updated on September 17, 2026.

Source: devcodex-labs/devcodex on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.