Ready to supercharge your AI agents?

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

Get Started Free →

Init Workspace Context · griddynamics | DecimalAI

agent toolingby griddynamics

Init Workspace Context

griddynamics/init-workspace-context

Source griddynamics/rosetta ↗License Apache-2.0

Classify workspace mode.

SkillScore82Maintained

On-demandOn-demand: invoked explicitly — never occupies your agent's context until called.

AI Engineer pack

Last verified 8/21/2026

What it changes

Passed 16 of 19 test cases with the skill, and 8 of 19 without it.

Without skill41%

With skill82%

Checks it passed that the model alone missed

Sets the composite flag to true
Identifies the workspace mode as upgrade
Identifies the workspace mode as install

and 8 more in the full report

9 fixed9 passed either way4 still failing0 broke

Pass-rate lift: +41 pp
Tokens: +63%
Agent turns: 0%
Response time: -39%

Model gemini-3.6-flashMeasured 8/21/2026v118 of 22 cases passed with skill · 19 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/griddynamics-init-workspace-context/SKILL.md https://app.decimal.ai/s/griddynamics-init-workspace-context/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add griddynamics-init-workspace-context, 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/griddynamics-init-workspace-context/ and pins it in .decimal/skills.lock

decimalai skills add griddynamics-init-workspace-context

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

curl -fsSL --create-dirs -o .claude/skills/griddynamics-init-workspace-context/SKILL.md https://app.decimal.ai/s/griddynamics-init-workspace-context/SKILL.md

Raw file: /s/griddynamics-init-workspace-context/SKILL.md ↗ · machine-readable summary at /s/griddynamics-init-workspace-context.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 9 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 9

Prompt

The LLM context has no active plugin flags. Scanning the root directory reveals two existing bootstrap configuration files. Which initialization mode applies?

What was checked, without and with the skill
Checked	Without skill	With skill
Identifies the workspace mode as upgrade	Missed	Passed

case-05 · gemini-3.6-flash · 8/21/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md93 linessingle file

<init_workspace_context>

<when_to_use_skill> Initialization must behave differently for fresh, existing, or plugin workspaces. Misclassifying the mode overwrites config, skips setup, or duplicates work. First skill in the init flow — runs before all others. Also generates or updates gain.json with SDLC tooling configuration collected from the user. </when_to_use_skill>

<core_concepts>

All Rosetta prep steps MUST be FULLY completed, load-context skill loaded and fully executed
Three modes: install (no files per bootstrap_rosetta_files), upgrade (some files per bootstrap_rosetta_files exist), plugin (LLM context already contains "RUNNING AS A PLUGIN")
gain.json is the single source of truth for SDLC tooling; it wins in conflicts with other files </core_concepts>

<gain_json_generation>

Auto-detect fields from workspace. Non-obvious: versions.rosetta — read from Rosetta plugin path in current LLM context, e.g. "2.0.43"; versions is for GAIN suite tools only (rosetta, specflow, etc.), not the project
Ask user for anything unresolved in a single batch. All fields optional. Prioritize critical fields first. Leave placeholders for skipped fields
If gain.json already exists: read it, ask only about gaps and placeholders; never overwrite confirmed values
If running as subagent: output gain_json_draft (template with auto-detected values filled in, all other fields left as template placeholders) to orchestrator; instruct orchestrator to ask user to fill in each placeholder field, and leave any unanswered fields exactly as the placeholder in the template

</gain_json_generation>

gain.json

Track record

Benchmark

16 / 19

test cases passed, vs 8 / 19 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

griddynamics / rosetta

instructions/r2/core/skills/init-workspace-context/SKILL.md

View on GitHub ↗

License Apache-2.0Imported 8/3/2026Version v1

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