Ready to supercharge your AI agents?

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

Compress a KB Prompt · griddynamics | DecimalAI

agent toolingby griddynamics

Compress a KB Prompt

griddynamics/compress-prompt

Source griddynamics/rosetta ↗License Apache-2.0

Compress a Rosetta KB prompt artifact (skill · workflow · phase · rule · agent · template · generic) by stripping structural tautology and ineffective scaffolding while preserving every importance-bearing token.

NewMaintained

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

AI Engineer pack

Last verified 8/27/2026

What it changes

Passed 3 of 10 test cases with the skill, and 4 of 10 without it.

Without skill41%

With skill32%

Checks it passed that the model alone missed

Invokes a low-reasoning Sonnet-5 class subagent model for execution review.
Appends a concise action item formatted under the self-improvement lessons learned section.
Restricts reads exclusively to the target artifact, its schema, and directive grammar, skipping unrelated repository docs.

3 fixed4 passed either way10 still failing5 broke

Pass-rate lift: -9 pp
Tokens: +199%
Agent turns: 0%
Response time: +28%

Model gemini-3.6-flashMeasured 8/27/2026v37 of 22 cases passed with skill · 10 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-compress-prompt/SKILL.md https://app.decimal.ai/s/griddynamics-compress-prompt/SKILL.md

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

decimalai skills add griddynamics-compress-prompt

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

curl -fsSL --create-dirs -o .claude/skills/griddynamics-compress-prompt/SKILL.md https://app.decimal.ai/s/griddynamics-compress-prompt/SKILL.md

Raw file: /s/griddynamics-compress-prompt/SKILL.md ↗ · machine-readable summary at /s/griddynamics-compress-prompt.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

I want you to compress `phases/auth_phase.md`. Before starting, should you read our general repository `README.md` and `docs/architecture_overview.md` to get context on the application?

What was checked, without and with the skill
Checked	Without skill	With skill
Restricts reads exclusively to the target artifact, its schema, and directive grammar, skipping unrelated repository docs.	Missed	Passed

case-17 · gemini-3.6-flash · 8/27/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md116 linessingle file

Compress a KB Prompt

Mental model

The artifact you compress is loaded into a coding agent that runs in ANOTHER repo, on ANOTHER user's task. Every token becomes that agent's context → scaffolding = distraction that dilutes focus.
Compress ≠ shrink words. Compress = strip scaffolding, keep 100% signal, sharpen focus, replace with meaningful unicode characters, like arrows.
Reader is AI like you, capable: it already knows the domain AND the KB grammar. Use terms and acronyms. Don't explain — nudge.
CAPS = importance. Word count is an OUTCOME, never a target.
INVARIANT ≠ STEP. Invariants (always-on constraints) → declare ONCE, flag always-on. Steps → ordered, run once. NEVER re-assert an invariant as per-step reminders — that's the echo authors reflexively add; compress hoists + cuts it.
GOLDEN RULE: NEVER trade a high-value token for a few saved words (unless it is repeatedly used and overall gives high results, we can loose 2% of value overall).
BULLETS vs ORDERED: ALWAYS convert bullets to ordered lists, if work is sequential or can be sequential. Reason: aligns with AI sequential token generation.
DENSIFY EVERY rule.
Take time to think during reasoning, take different options, iterate multiple times TRANSFORM.
Your task is MAXIMUM compression, not just low hanging fruits!
NO rush, TAKE time
Don't bring this skill terms or meta-thinking
Identify what is load bearing

Allowed reads — read-only, NEVER adjust

Read ONLY: the target artifact + its type schema + the grammar below. Nothing else. Stay focused.

Schemas (learn which XML scopes are MANDATORY vs optional, and each scope's role):

docs/schemas/skill.md
docs/schemas/workflow.md
docs/schemas/phase.md
docs/schemas/rule.md
docs/schemas/agent.md
docs/schemas/template.md
docs/schemas/generic.md

Grammar — directive commands the system ACTS ON; protect verbatim + their args:

Command	Semantics
`USE SKILL <name>` / `READ SKILL <name>`	activate skill (load `SKILL.md` + act) / load content only
`READ SKILL FILE <subpath>` / `APPLY SKILL FILE <subpath>`	load / load+execute a file of the CURRENT skill; never names a skill (isolation is grammar-enforced)
`USE FLOW <name>.md` / `READ FLOW <name>.md`	invoke a whole workflow / load without executing
`APPLY PHASE <file>.md`	load + fully execute the next phase body of a running workflow
`INVOKE SUBAGENT <name>` / `READ SUBAGENT <name>`	spawn subagent / load its definition only
`READ RULE <file>.md` / `APPLY RULE <file>.md`	load / load+execute a rule
`READ TEMPLATE <file>.md`	load a template

Track record

Benchmark

3 / 10

test cases passed, vs 4 / 10 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 v5 — 8/26/2026

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

.claude/skills/compress-prompt/SKILL.md

View on GitHub ↗

License Apache-2.0Imported 8/26/2026Version v3

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

Compress a KB Prompt

Mental model

Allowed reads — read-only, NEVER adjust

Track record

Benchmark

Adoption

SkillSafety

Source

Related skills