Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add JetXu-LLM/DocMason --skill knowledge-base-syncgit clone --depth 1 https://github.com/JetXu-LLM/DocMasonWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/jetxu-llm/docmason/knowledge-base-sync)<a href="https://agentmods.dev/skills/jetxu-llm/docmason/knowledge-base-sync"><img src="https://agentmods.dev/badge/skills/jetxu-llm/docmason/knowledge-base-sync/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/jetxu-llm/docmason/knowledge-base-sync"><img src="https://agentmods.dev/badge/skills/jetxu-llm/docmason/knowledge-base-sync.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00026 | $0.01227 |
| Opus 5.5 | $0.00010 | $0.00491 |
| Sonnet 5.5 | $0.00005 | $0.00245 |
| Haiku 4.5 | $0.00003 | $0.00123 |
Grade A, and why
knowledge-base-sync scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Knowledge-Base Sync
Use this skill when the task is to build or refresh the DocMason knowledge base.
This is a top-level operator workflow.
Ordinary users should not need to name knowledge-construction or validation-repair themselves.
Required Capabilities
- local file access
- shell or command execution
- ability to inspect structured JSON output
If the agent cannot run local commands or inspect the resulting artifacts, stop and explain that the workspace cannot be synced reliably.
Procedure
- Start with
docmason status --jsonwhen you need the current stage, pending actions, or control-plane state. - Use
docmason sync --jsonas the default build or refresh entry point.- let sync classify the physical cause before staging
zero-deltaand foregroundinteraction-promotion-onlychecks return the current legal truth without full staging, validation, or publication; pending interaction promotion stays deferred maintenance rather than an answer-critical invalidation- for a source delta, preserve source-scoped transactional staging, cached validation for unchanged sources, compact global checks, and the existing atomic single-current publish model; do not improvise a parallel full-copy or publication path
- use the reported cause, dirty/reused counts, phase timings, cache hits/misses, and physical clone/copy telemetry when diagnosing performance or unexpected work
- keep workspace commands sequential while the sync lease is active
- do not overlap
status,sync,retrieve,trace, orvalidate-kbagainst the same workspace
- Respect the control-plane result before reasoning about later workflow steps.
- if
sync_status=awaiting-confirmation, surface the confirmation prompt and continue withdocmason sync --yes --json - if
sync_status=waiting-shared-job, treat the existing shared sync job as the legal owner and wait or retry rather than starting a second path - if
sync_status=action-required, surface the blocker directly- when the blocker is missing sync capability, route the operator to
prepare - when the blocker is a repairable Office machine-baseline gap and the host can provide
Full access, continue throughdocmason prepare --yes --jsonand then resume the same sync task instead of stopping at a passive blocker report
- when the blocker is missing sync capability, route the operator to
- if
- Treat successful
syncas the deterministic truth-building path.- detect source changes
- rebuild or reuse staged evidence
- apply safe staged repairs
- validate
- publish to
knowledge_base/current
- If
sync_statusisvalidorwarnings, inspecthybrid_enrichment.hybrid_enrichmentdescribes whether deterministic sync left a remaining multimodal semantic gap- if
modeisnot-neededorcovered, stop and return the publication result - if the sync payload also includes
lane_b_follow_up.work_path, open that governed work packet first and treat it as the authoritative next step for this sync statelane_b_follow_upis the bounded sync-time follow-up packet for that current staging state
- when that governed packet is present, do not consume the broad queue blindly; only fall back to
hybrid_work_pathwhen no bounded packet was handed offhybrid_work_pathis the broader staged queue and is only the fallback when no bounded packet was handed off
- if
modeiscandidate-preparedorpartially-coveredand no governed packet path was handed off, treathybrid_work_pathas the authoritative hard-artifact queue - do not improvise a second sync path or rewrite deterministic sidecars
- write only additive
semantic_overlay/sidecars for units you can support honestly, then rerundocmason sync --json
- Treat
pending-synthesisas a compatibility or deliberate manual-mode state, not the normal operator destination.- route to
knowledge-constructiononly when that legacy or manual path is actually intended - rerun
docmason sync --jsonafter staged authoring completes
- route to
- If validation still blocks publication after the deterministic sync path finishes, route to
validation-repair, then rerundocmason sync --json. - Return the final publication judgment to the main agent. Do not delegate final publication sign-off.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 81 lines · 26 tokens per session scan A aa1df4accae2
knowledge-base-sync is a skill published in the GitHub repository JetXu-LLM/DocMason (150 stars, last pushed 3d ago), licensed Apache-2.0. It adds 26 tokens to every session and 1,227 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
transformer-attention
Use when reasoning about Transformer self-attention, multi-head attention, positional encoding, masked decoder attention, or why attention replaced recurrence/convolutions in sequence models; not for generic NLP or unrelated attention topics.
thinking-out-loud
A contract for what the agent does when a long, messy, stream-of-consciousness ramble arrives (usually voice dictation): act on nothing until the echo brief is approved. The echo audits the entire transfer, mission, locked decisions and constraints, open questions, flips and parked tangents, with the model's…
hugging-face-datasets
Create and manage datasets on Hugging Face Hub. Supports initializing repos, defining configs/system prompts, streaming row updates, and SQL-based dataset querying/transformation. Designed to work alongside HF MCP server for comprehensive dataset workflows.
azure-search-documents-dotnet
Azure AI Search SDK for .NET (Azure.Search.Documents). Use for building search applications with full-text, vector, semantic, and hybrid search. Covers SearchClient (queries, document CRUD), SearchIndexClient (index management), and SearchIndexerClient (indexers, skillsets). Triggers: "Azure Search .NET"…
azure-ai-ml-py
Azure Machine Learning SDK v2 for Python. Use for ML workspaces, jobs, models, datasets, compute, and pipelines. Triggers: "azure-ai-ml", "MLClient", "workspace", "model registry", "training jobs", "datasets".
capacity
Discovers available Azure OpenAI model capacity across regions and projects. Analyzes quota limits, compares availability, and recommends optimal deployment locations based on capacity requirements. USE FOR: find capacity, check quota, where can I deploy, capacity discovery, best region for capacity, multi-project…