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 caipe-io/ai-platform-engineering --skill documentation-searchgit clone --depth 1 https://github.com/caipe-io/ai-platform-engineeringWrote 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/caipe-io/ai-platform-engineering/documentation-search)<a href="https://agentmods.dev/skills/caipe-io/ai-platform-engineering/documentation-search"><img src="https://agentmods.dev/badge/skills/caipe-io/ai-platform-engineering/documentation-search.svg" alt="Measured on agentmods" 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 | $0.00046 | $0.00964 |
| Opus 5 | $0.00023 | $0.00482 |
| Sonnet 5 | $0.00009 | $0.00193 |
| Haiku 4.5 | $0.00005 | $0.00096 |
Grade A, and why
documentation-search 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 8d 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 — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Documentation Search
Search the internal knowledge base using RAG (Retrieval-Augmented Generation) to find relevant runbooks, architecture docs, ADRs, and best practices.
Instructions
Phase 1: Query Understanding
- Parse the user's intent:
- Are they looking for a specific document (e.g., "deployment runbook")?
- Are they asking a question the docs can answer (e.g., "how do we handle rollbacks?")?
- Are they exploring a topic (e.g., "what do we know about A2A protocol?")?
- Formulate search queries:
- Extract key terms from the user's question
- Generate 2-3 variant queries to improve recall
- Include relevant synonyms (e.g., "deploy" = "release" = "ship")
Phase 2: Knowledge Base Search (RAG Agent)
- Search across document sources:
- Architecture Decision Records (ADRs) in
docs/docs/changes/ - Spec Kit documents in
.specify/specs/ - Runbooks and operational guides
- README files and inline documentation
- Confluence pages (if Confluence agent available)
- Backstage TechDocs (if Backstage agent available)
- Architecture Decision Records (ADRs) in
- Rank results by relevance:
- Exact keyword matches ranked highest
- Semantic similarity for conceptual matches
- Recency as a tiebreaker (newer docs preferred)
Phase 3: Answer Synthesis
- Direct answer: If the docs contain a clear answer, provide it directly
- Compiled answer: If information is spread across multiple docs, synthesize
- Source attribution: Always cite which document(s) the answer came from
- Gaps identified: Note if the question is only partially answered
Output Format
## Documentation Search Results
**Query**: "How do we handle database migrations?"
### Answer
Based on the internal documentation, database migrations follow this process:
1. Create a migration script using Alembic (see Runbook RB-023)
2. Test in staging environment first (required per ADR-008)
3. Run during the maintenance window defined in the on-call calendar
4. Verify with rollback script before marking complete
### Sources
| Document | Type | Relevance | Last Updated |
|----------|------|-----------|--------------|
| RB-023: Database Migration Runbook | Runbook | High | 2026-01-15 |
| ADR-008: Database Schema Changes | ADR | Medium | 2025-11-20 |
| .specify/specs/db-migration-v2.md | Spec | Medium | 2026-02-01 |
### Key Excerpts
> From **RB-023**: "Always run `alembic upgrade head --sql` first to preview
> the migration SQL before applying. Never run migrations during peak hours."
> From **ADR-008**: "We chose Alembic over Django migrations because our
> services are not Django-based and Alembic provides better raw SQL support."
### Related Topics
- Rollback procedures (see RB-024)
- Schema versioning strategy (see ADR-008)
- Testing database changes (see .specify/specs/db-testing.md)
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.
- 8d ago First seen · 98 lines · 46 tokens per session scan A fcfbcc542ad9
documentation-search is a skill published in the GitHub repository caipe-io/ai-platform-engineering (407 stars, last pushed today), licensed Apache-2.0. It adds 46 tokens to every session and 964 once invoked, about $0.0002 per session on Opus 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
mem0-integration
Mem0 memory layer integration for AI agents. Implement persistent, semantic memory for long-term context retention and personalization.
vector-memory
HNSW vector search for pattern similarity retrieval and knowledge graph maintenance with PageRank scoring, community detection, and 3-tier memory management.
chroma-integration
Chroma local vector database setup and operations for development and production.
haystack-pipeline
Haystack NLP pipeline configuration for document processing and QA.
langchain-retriever
LangChain retriever implementation with various retrieval strategies for RAG applications.
llamaindex-agent
LlamaIndex agent and query engine setup for RAG-powered agents.