Ready to supercharge your AI agents?

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

Knowledge Layer Skill — measured AI agent skill | DecimalAI

study8677/knowledge-layer

NewSkillScoreMaintained

High-level deployment wrapper over RepoBrain core with graph-first knowledge injection and all-file support. Exposes `refresh_filesystem` and `ask_filesystem` for building and querying the knowledge graph.

Documentationby study8677

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Knowledge Management pack

decimalai skills pull study8677-knowledge-layer

Get this skill — curl · raw file

curl -fsSL https://app.decimal.ai/s/study8677-knowledge-layer/SKILL.md

Raw file: /s/study8677-knowledge-layer/SKILL.md ↗ · machine-readable summary at /s/study8677-knowledge-layer.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 3h ago

Yes — +59% lift-53% tokens0% turns86% of whole cases pass with skill (19/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-04	✗→✓	▲ Improved	-64%	0%
case-06	✗→✓	▲ Improved	-28%	0%
case-07	✗→✓	▲ Improved	-68%	0%
case-09	✗→✓	▲ Improved	-72%	0%
case-10	✗→✓	▲ Improved	-46%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

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

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

study8677 / repobrain

engine/repobrain_engine/skills/knowledge-layer/SKILL.md

View on GitHub ↗

License MITImported 8/3/2026Version v1

Skill contents

SKILL.md

Knowledge Layer Skill

Purpose

Provide a high-level deployment wrapper over RepoBrain core, with graph-first knowledge injection and all-file support (code, docs, data, media metadata).

Inputs

refresh_filesystem(workspace=".", quick=False)
ask_filesystem(question, workspace=".")

Outputs

Refresh writes graph-first artifacts under .repobrain/:
knowledge_graph.json
knowledge_graph.md
knowledge_graph.mmd
document_index.md
data_overview.md
media_manifest.md
plus existing conventions.md and structure.md
Ask returns a grounded answer with source paths.

Boundaries

Skill is a wrapper layer only; no standalone runtime.
Core Hub/Agent/Pipeline architecture remains the source of truth.

Compatibility

Existing commands (rb-refresh, rb-ask, rb-mcp) remain valid.
High-level aliases can be disabled with RB_ENABLE_LAYER_ALIASES=0.

Degrade Strategy

If graph artifacts are unavailable, ask falls back to structure.md and

conventions.md context.

Want to see if study8677/knowledge-layer would activate on your phrasing? Try it before you install — no LLM call, no signup.

Other measured skills in the registry, with their headline benchmark lift.