Ready to supercharge your AI agents?

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

Dev Deep Dive · evolution-foundation | DecimalAI

researchby evolution-foundation

Dev Deep Dive

evolution-foundation/dev-deep-dive

Source evolution-foundation/evo-nexus ↗License NOASSERTION

Two-stage investigation — causal trace (via @trail-tracer) followed by requirements crystallization (via @echo-analyst + dev-deep-interview).

SkillScore100Maintained

Automatic

Coding pack

Last verified 8/19/2026

What it changes

Passed 22 of 22 test cases with the skill, and 20 of 22 without it.

Without skill91%

With skill100%

Checks it passed that the model alone missed

Addresses dark mode styling criteria directly as requested
Generates CSS component tasks directly without initiating a diagnostic workflow
Establishes partial index freshness targets and resource limits prior to fix planning

2 fixed20 passed either way0 still failing0 broke

Pass-rate lift: +9 pp
Tokens: +23%
Agent turns: 0%
Response time: +10%

Model gemini-3.6-flashMeasured 8/19/2026v122 of 22 cases passed with skill

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/evolution-foundation-dev-deep-dive/SKILL.md https://app.decimal.ai/s/evolution-foundation-dev-deep-dive/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add evolution-foundation-dev-deep-dive, 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/evolution-foundation-dev-deep-dive/ and pins it in .decimal/skills.lock

decimalai skills add evolution-foundation-dev-deep-dive

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

curl -fsSL --create-dirs -o .claude/skills/evolution-foundation-dev-deep-dive/SKILL.md https://app.decimal.ai/s/evolution-foundation-dev-deep-dive/SKILL.md

Raw file: /s/evolution-foundation-dev-deep-dive/SKILL.md ↗ · machine-readable summary at /s/evolution-foundation-dev-deep-dive.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 2 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 2

Prompt

The design team provided a complete specification with explicit criteria for adding dark mode support to all account settings screens, and there are no backend errors or unknown bugs. Please generate the CSS component implementation tasks directly.

What was checked, without and with the skill
Checked	Without skill	With skill
Generates CSS component tasks directly without initiating a diagnostic workflow	Missed	Passed
Addresses dark mode styling criteria directly as requested	Missed	Passed

case-05 · gemini-3.6-flash · 8/19/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md38 linessingle file

Dev Deep Dive

Derived from oh-my-claudecode (MIT, Yeachan Heo). Adapted for the EvoNexus Engineering Layer.

A two-stage investigation that combines causal tracing with requirements crystallization. Use when both the cause AND the scope of a problem are ambiguous.

Use When

Production issue with unclear cause AND unclear scope of fix
Bug report that needs both root cause analysis and clarification of "what does fixed look like?"
Strategic problem where you don't know what to build OR why the current state isn't working

Workflow

Stage 1 — Causal trace

Hand off to @trail-tracer:

Generate competing hypotheses
Collect evidence FOR and AGAINST each
Apply lenses (systems / premortem / science)
Identify the discriminating probe
Save report to workspace/development/debug/[C]trace-{topic}-{date}.md

Stage 2 — Requirements crystallization

Hand off to @echo-analyst + dev-deep-interview:

Surface unstated assumptions
Define scope boundaries
Identify missing acceptance criteria
Save spec to workspace/development/specs/[C]spec-{topic}-{date}.md

Stage 3 — Synthesis

Combine the trace report and the requirements spec
Output: a unified understanding of "what's broken AND what done looks like"
Hand off to @compass-planner to plan the actual fix

Pairs With

@trail-tracer (Stage 1)
@echo-analyst (Stage 2)
dev-deep-interview (Stage 2)
@compass-planner (Stage 3 handoff)

Track record

Benchmark

22 / 22

test cases passed, vs 20 / 22 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

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

evolution-foundation / evo-nexus

.claude/skills/dev-deep-dive/SKILL.md

View on GitHub ↗

License NOASSERTIONImported 8/3/2026Version v1

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

Ready to supercharge your AI agents?

See it in action

What's inside

Dev Deep Dive

Use When

Workflow

Stage 1 — Causal trace

Stage 2 — Requirements crystallization

Stage 3 — Synthesis

Pairs With

Track record

Benchmark

Adoption

SkillSafety

Source

Related skills