Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Project Incarnation · linsir2 | DecimalAI

agent toolingby linsir2

Project Incarnation

linsir2/project-incarnation

Source linsir2/project-incarnation-skill ↗License MIT

项目化身术：把 repo / 技术栈 / 方法论 / 现有 skill 蒸馏成一个高知识密度、可直接运行的 `SKILL.md`。目标不是产出 bundle，而是产出“技术化身”：能复现对象的判断、边界、偏好、反模式与求证路径。适用于：蒸馏项目、框架、SDK、工程体系、已有 skill。

SkillScore77Maintained

Automatic

AI Engineer pack

Last verified 8/3/2026

What it changes

Passed 15 of 20 test cases with the skill, and 12 of 20 without it.

Without skill59%

With skill77%

Checks it passed that the model alone missed

The decision crystal attaches evidence level, time sensitivity, exclusivity, and confidence metrics to the Rust memory safety rules.
The document structures variable guidance around domain/organizational context boundaries rather than software package version numbers.
The update preserves the existing SKILL.md structure and stable core, updating only version scope, research date, and version-sensitive crystals.

and 3 more in the full report

6 fixed11 passed either way3 still failing2 broke

Pass-rate lift: +18 pp
Tokens: +248%
Agent turns: 0%
Response time: +22%

Model gemini-3.6-flashMeasured 8/3/2026v117 of 22 cases passed with skill · 20 comparable

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/linsir2-project-incarnation/SKILL.md https://app.decimal.ai/s/linsir2-project-incarnation/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add linsir2-project-incarnation, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/linsir2-project-incarnation/ and pins it in .decimal/skills.lock

decimalai skills add linsir2-project-incarnation

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/linsir2-project-incarnation/SKILL.md https://app.decimal.ai/s/linsir2-project-incarnation/SKILL.md

Raw file: /s/linsir2-project-incarnation/SKILL.md ↗ · machine-readable summary at /s/linsir2-project-incarnation.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 6 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 6

Prompt

生成 SKILL.md 之后需要进行哪四种特定验证测试／请列出这四种测试方法的名称。

What was checked, without and with the skill
Checked	Without skill	With skill
The response details four validation methodologies covering replay testing, extrapolation, boundary testing, and isolated single-file verification.	Missed	Passed

case-16 · gemini-3.6-flash · 8/3/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md777 linessingle file

项目化身术

不做文档复印机，只做判断压缩机。

核心理念

项目化身不是“介绍这个东西是什么”，而是把一个对象压缩成一份可继续做判断的认知器官。

一个合格的技术化身，至少要让后续的 agent / harness / 人类读者只靠这一份 SKILL.md 就知道：

这个对象到底在保护什么真相
它最常反对什么错误
它遇到新问题时先看什么证据
它在哪些地方必须闭嘴或降置信度
它的判断为什么不是通用废话，而是这个对象独有的判断

关键区分：

不是复制文档，而是提炼判断
不是列功能点，而是提炼取舍
不是写“它很厉害”，而是写“它会如何反对你”
不是生成一堆附属文件，而是生成一个自包含的 SKILL.md

最终产物只有：

<runtime-skill-root>/<slug>/
└── SKILL.md

其中 <runtime-skill-root> 指的是agent 运行时可直接发现和加载的安装目录，例如：

<project-root>/.agents/skills/
<project-root>/.codex/skills/
$CODEX_HOME/skills/
~/.codex/skills/

不生成 constitution.yaml、sources.jsonl、bundle-spec.json、evals.md、references/、scripts/。

Track record

Benchmark

15 / 20

test cases passed, vs 12 / 20 without

Adoption

Users: 1
Times used: 3

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/1/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

linsir2 / project-incarnation-skill

SKILL.md

View on GitHub ↗

License MITImported 8/1/2026Version v1

More skills from the same source and nearby in the registry, with their headline benchmark lift.